A dating and friendship matching method and system based on deep learning neural network

Abstract

The present application discloses a dating and making friends matching method and system based on a deep learning neural network. First, the gender information and facial image of the user to be matched are obtained, and facial feature data is obtained based on the facial image of the user to be matched; then the gender information and facial feature data of the user to be matched are input into a couple matching mathematical model to obtain couple matching feature data, and the couple matching feature data is matched with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the couple matching feature data; finally, the opposite sex information corresponding to the target facial feature data is output. Since the gender information and facial image of the user to be matched are input into the couple matching mathematical model to obtain the couple matching feature data, and matching is performed based on the couple matching feature data, the satisfaction of both parties in the dating and making friends at the first meeting is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of blind date and friendship intelligent matching, and in particular to a blind date and friendship matching method and system based on a deep learning neural network. Background Art

[0002] The first meeting between two blind dates is awkward, and it is also a waste of time if you don't know each other very well. How to improve the success rate of blind dates is a matter of great concern to young people of marriageable age, matchmaking agencies, and parents. The first impression is the most important in the process of strangers getting to know each other, and it is usually obtained through photos or descriptions of introducers, but a lot of information is missing. In particular, many pictures used for blind dates are processed, which leads to the discovery that they are far from their first impression when they meet face to face, wasting both parties' time and energy. How to ensure the satisfaction of both parties in the first meeting is a prerequisite for improving the success rate of blind date matching. Summary of the invention

[0003] The main technical problem solved by the present invention is how to improve the satisfaction of both parties in a blind date or friendship meeting for the first time.

[0004] According to the first aspect, an embodiment provides a dating and friendship matching method based on a deep learning neural network, comprising:

[0005] Obtain the gender information and facial image of the user to be matched;

[0006] Acquiring facial feature data based on the facial image of the user to be matched;

[0007] Inputting the gender information and facial feature data of the user to be matched into a couple matching mathematical model, and obtaining the couple matching feature data output by the couple matching mathematical model; the couple matching feature data is conditional data that meets the couple appearance of the user to be matched;

[0008] The matching feature data of the couple are matched with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the matching feature data of the couple, and output opposite-sex information corresponding to the target facial feature data; the target facial database is used to store the opposite-sex facial data to be matched.

[0009] In one embodiment, the step of acquiring facial feature data based on the facial image of the user to be matched includes:

[0010] Performing face detection on the facial image and acquiring a facial image;

[0011] Preprocessing the facial image; the preprocessing includes scale normalization and / or grayscale value correction;

[0012] Acquire positioning feature points of the preprocessed face image; the positioning feature points are used to locate the face image, and the positioning feature points include eyes, ears, nose, chin, corners of mouth and / or center points of mouth;

[0013] The facial feature data is acquired according to the positioning feature points.

[0014] In one embodiment, the facial feature data includes face shape feature data, eye shape feature data, nose shape feature data and / or mouth shape feature data;

[0015] The facial feature data include face length, face width, forehead width, forehead height, forehead-zygomatic height difference, jaw width, jaw height, left frontal angle and / or right frontal angle;

[0016] The eye shape feature data include the width of the outer eyelids of both eyes, the width of the inner eyelids of both eyes, the width of the inner and outer palpebral fissures of the left eye, the width of the left palpebral fissure, the height difference between the inner and outer eyes of the left eye, the outer palpebral fissure angle of the left eye, the inner palpebral fissure angle of the left eye, the width of the inner and outer palpebral fissure of the right eye, the width of the right palpebral fissure, the height difference between the inner and outer eyes of the right eye, the outer palpebral fissure angle of the right eye, the inner palpebral fissure angle of the right eye, the roundness of the left eye and / or the three-angle of the right eye, wherein the roundness of the left eye includes the polygonal area of ​​the left eye contour and / or the area of ​​the circumscribed circle of the left eye, and the three-angle of the right eye includes the area of ​​the inscribed triangle of the right eye and / or the polygonal area of ​​the right eye contour;

[0017] The nose shape feature data includes nose length, nose width, height difference between nose tip and nose base, right nose wing width, height difference between left nostril and nose base, left nostril width and / or left nostril height;

[0018] The mouth shape feature data include lip width, lip height, width from the midpoint of the upper lip to the left lip peak, width from the midpoint of the upper lip to the right lip peak, height from the midpoint of the upper lip to the left lip peak, height from the midpoint of the upper lip to the right lip peak, height from the left corner of the mouth to the left lip peak, height from the left corner of the mouth to the lip bead, height from the right corner of the mouth to the lip bead, height from the middle of the upper lip to the lip bead and / or height from the lip bead to the midpoint of the lower lip.

[0019] In one embodiment, the facial feature data includes face shape, eye shape, nose shape and / or mouth shape;

[0020] The eye shapes include droopy eyes, phoenix eyes, hanging eyes, near-centered eyes, squinting eyes, triangular eyes, elongated eyes, almond eyes, round eyes and / or telecentric eyes;

[0021] The nose shapes include upturned nose, short flat nose, Nubian nose, Greek nose and / or hooked nose;

[0022] The mouth shapes include the boat-shaped mouth, the bow-shaped mouth, the tiger's mouth, the four-character mouth, the upturned moon mouth and / or the cherry mouth;

[0023] The face shapes include round, square, heart, oblong, diamond, oval and / or pear.

[0024] In one embodiment, the method for obtaining a mathematical model for matching a couple includes:

[0025] Acquire a preset number of facial images of a couple whose marriage age meets a preset time period;

[0026] Extracting facial feature data of the husband and wife respectively based on the facial images of the husband and wife;

[0027] The facial feature data of the couple are used as elements in a training set to train the couple matching mathematical model to obtain the corresponding relationship between the facial feature data and the couple matching feature data.

[0028] In one embodiment, matching the matching feature data of the couple with a preset target facial database includes:

[0029] The opposite-sex facial data to be matched includes the opposite-sex facial image to be matched;

[0030] The LBP operator is used to compare the matching feature data of the couple with the face images of the opposite sex in the target face database.

[0031] In one embodiment, matching the matching feature data of the couple with a preset target facial database includes:

[0032] The opposite-sex facial data to be matched includes facial feature data of the opposite-sex to be matched;

[0033] The matching feature data of the couple is matched with the facial feature data of the opposite sex to be matched.

[0034] According to the second aspect, an embodiment provides a computer-readable storage medium, including a program, wherein the program can be executed by a processor to implement the blind date and friendship matching method described in the first aspect.

[0035] According to the third aspect, an embodiment provides a dating and friendship matching system based on a deep learning neural network, including a dating matching server, which is used to obtain gender information and facial feature data of a user to be matched; the facial feature data is obtained based on a facial image of the user to be matched;

[0036] The matchmaking server is further used to input the gender information and facial feature data of the to-be-matched user into a couple matching mathematical model, and obtain the couple matching feature data output by the couple matching mathematical model; the couple matching feature data is conditional data that meets the couple appearance of the to-be-matched user;

[0037] The blind date matching server is also used to match the matching feature data of the couple with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the matching feature data of the couple, and output opposite-sex information corresponding to the target facial feature data; the target facial database is used to store the opposite-sex facial data to be matched.

[0038] In one embodiment, the matchmaking and dating matching system further includes a matchmaking matching terminal for obtaining the gender information and facial image of the user to be matched;

[0039] The blind date matching terminal is also used to obtain facial feature data based on the facial image of the user to be matched, and send the gender information and facial feature data of the user to be matched to the blind date matching server;

[0040] And / or, the blind date matching terminal is further used to obtain the opposite-sex information corresponding to the target facial feature data output by the blind date matching server and display it.

[0041] According to the above-mentioned embodiment, a blind date and friendship matching method based on deep learning neural network inputs the gender information and facial image of the user to be matched into the couple matching mathematical model to obtain couple matching feature data, and matches are performed based on the couple matching feature data, thereby improving the satisfaction of both parties in the blind date and friendship when they meet for the first time. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A schematic diagram of a process flow of a dating and making friends matching method in an embodiment;

[0043] Figure 2 Diagram of eye shape types;

[0044] Figure 3 Diagram of nose shape types;

[0045] Figure 4 Diagram of mouth shape types;

[0046] Figure 5 Diagram of face shape types;

[0047] Figure 6 A schematic diagram of facial feature data in an embodiment;

[0048] Figure 7 A schematic diagram of eye shape feature data in an embodiment;

[0049] Figure 8 A schematic diagram of nose shape feature data in an embodiment;

[0050] Fig. 9A schematic diagram of mouth shape feature data in an embodiment;

[0051] Fig.10 It is a structural diagram of a blind date and friendship matching system in another embodiment. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0053] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0054] The serial numbers assigned to the components herein, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings).

[0055] At present, the blind date and friendship matching system searches based on the blind date and friendship conditions pre-entered by the person to be matched, and then selects based on the photos retrieved from the numerous IDs. This method is not only inefficient, but also causes unnecessary embarrassment when meeting due to the difference between the real person and the photo. In addition, since the photo of the person to be matched can be directly obtained, the privacy of the person to be matched who does not meet the meeting conditions cannot be protected. In the embodiment of the present application, the gender information and facial image of the user to be matched are input into the mathematical model for matching couples to obtain the matching feature data of the couple, and matching is performed based on the matching feature data of the couple, so that the satisfaction of the two parties in the blind date and friendship meeting for the first time is improved. In addition, since the matching is only performed based on the facial feature data of the two parties in the blind date and friendship, the photos of the person to be matched will not be made public, so that the privacy of the person to be matched who does not meet the meeting conditions is protected.

[0056] Embodiment 1:

[0057] Please refer to Figure 1 , is a flow chart of a blind date and friend matching method in an embodiment, the blind date and friend matching method comprising:

[0058] Step 100, obtaining gender information and facial image.

[0059] Get the gender information and facial image of the user to be matched.

[0060] Step 200, obtaining facial feature data.

[0061] Obtain facial feature data based on the facial image of the user to be matched, including:

[0062] First, face detection is performed on the facial image and a facial image is obtained. A facial image is captured from the facial image of the user to be matched.

[0063] Then the face image is preprocessed, and general preprocessing includes scale normalization and grayscale value correction.

[0064] Then, the positioning feature points of the preprocessed face image are obtained, and the positioning feature points are used to locate the face image. Among them, the positioning feature points include eyes, ears, nose, chin, mouth corners and the center point of the mouth, etc. The face is located in the plane through the center point of the facial features.

[0065] Finally, facial feature data is obtained based on the positioning feature points. In one embodiment, the facial feature data includes face shape, eye shape, nose shape and mouth shape. Among them, eye shapes include drooping eyes, phoenix eyes, hanging eyes, near-centered eyes, squinting eyes, triangular eyes, slender eyes, almond eyes, round eyes and telecentric eyes; nose shapes include upturned noses, short flat noses, Nubian noses, Greek noses and hooked noses; mouth shapes include boat-shaped mouths, arched mouths, tiger mouths, four-character mouths, upturned moon mouths and cherry mouths; face shapes include round, square, heart-shaped, long, diamond-shaped, egg-shaped and pear-shaped. Please refer to Figure 2 , Figure 3 , Figure 4 and Figure 5 , which are schematic diagrams of eye shape, nose shape, mouth shape and face shape.

[0066] In one embodiment, the facial feature data includes face shape feature data, eye shape feature data, nose shape feature data and mouth shape feature data.

[0067] Please refer to Figure 6 , is a schematic diagram of facial feature data in an embodiment, the facial feature data includes face length X1, face width X2, forehead width X3, forehead height X4, forehead and cheekbone height difference X5, jaw width X6, jaw height X7, left frontal angle X8 and right frontal angle X9.

[0068] Please refer to Figure 7 , is a schematic diagram of eye shape feature data in an embodiment, wherein the eye shape feature data includes the width of the outer eyelids of both eyes Y1, the width of the inner eyelids of both eyes Y2, the width of the inner and outer palpebral fissures of the left eye Y3, the width of the left palpebral fissure Y4, the height difference between the inner and outer eyes of the left eye Y5, the outer palpebral fissure angle of the left eye Y6, the inner palpebral fissure angle of the left eye Y7, the width of the inner and outer palpebral fissure of the right eye Y8, the width of the right palpebral fissure Y9, the height difference between the inner and outer eyes of the right eye Y10, the outer palpebral fissure angle of the right eye Y11, the inner palpebral fissure angle of the right eye Y12, the roundness of the left eye Y13, and the three-angle of the right eye Y14. Among them, the roundness of the left eye Y13 includes the area of ​​the left eye contour polygon or the area of ​​the circumscribed circle of the left eye, and the three-angle of the right eye Y14 includes the area of ​​the inscribed triangle of the right eye or the area of ​​the right eye contour polygon.

[0069] Please refer to Figure 8 , is a schematic diagram of nose shape feature data in an embodiment, the nose shape feature data includes nose length B1, nose width B2, height difference between nose tip and nose base B3, right nose wing width B4, height difference between left nostril and nose base B5, left nostril width B6 and left nostril height B7.

[0070] Please refer to Fig. 9 , is a schematic diagram of mouth shape feature data in an embodiment, the mouth shape feature data include lip width Z1, lip height Z2, width from the middle point of the upper lip to the left lip peak Z3, width from the middle point of the upper lip to the right lip peak Z4, height from the middle point of the upper lip to the left lip peak Z5, height from the middle point of the upper lip to the right lip peak Z6, height from the left corner of the mouth to the left lip peak Z7, height from the left corner of the mouth to the lip bead Z8, height from the right corner of the mouth to the lip bead Z9, height from the middle of the upper lip to the lip bead Z10 and height from the lip bead to the middle point of the lower lip Z11.

[0071] Step 300, input the couple matching data model.

[0072] The gender information and facial feature data of the user to be matched are input into a mathematical model for matching a couple. The feature data for matching a couple is conditional data that meets the condition that the user to be matched looks like a couple. In one embodiment, the method for obtaining the mathematical model for matching a couple includes:

[0073] Acquire a preset number of facial images of a couple whose marriage age meets a preset time period;

[0074] Extracting facial feature data of the husband and wife respectively based on their facial images;

[0075] The facial feature data of the couple are used as elements in the training set to train the couple matching mathematical model to obtain the corresponding relationship between the facial feature data and the couple matching feature data.

[0076] Step 400, obtaining matching characteristic data of the couple.

[0077] The couple matching feature data output by the couple matching mathematical model is obtained.

[0078] Step 500, matching is performed in the target face database.

[0079] The acquired matching feature data of the couple is matched with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the matching feature data of the couple. The target facial database is used to store the facial data of the opposite sex to be matched.

[0080] In one embodiment, the opposite-sex facial data to be matched includes opposite-sex facial images to be matched, and the acquired matching feature data of the couple is matched with a preset target facial database by applying an LBP operator to compare the matching feature data of the couple with the opposite-sex facial images in the target facial database.

[0081] In one embodiment, the facial data of the opposite sex to be matched includes facial feature data of the opposite sex to be matched, and matching the acquired matching feature data of the couple with a preset target facial database is matching the matching feature data of the couple with the facial feature data of the opposite sex to be matched.

[0082] Step 600, outputting the opposite-sex information that meets the matching conditions.

[0083] Output the opposite sex information corresponding to the target facial feature data.

[0084] In one embodiment of the present application, a dating and friendship matching system based on deep learning neural network is also disclosed. Please refer to Fig.10, is a structural diagram of a matchmaking and dating matching system in another embodiment, the matchmaking and dating matching system includes a matchmaking and dating matching server 20 and a matchmaking and dating matching terminal 10. The matchmaking and dating matching server 20 is used to obtain the gender information and facial feature data of a user to be matched, wherein the facial feature data is obtained based on the facial image of the user to be matched. The matchmaking and dating matching server 20 is also used to input the gender information and facial feature data of the user to be matched into a mathematical model for matching a couple, and obtain the couple matching feature data output by the mathematical model for matching a couple, and the couple matching feature data is the conditional data that meets the couple appearance of the user to be matched. The matchmaking and dating matching server 20 is also used to match the couple matching feature data with a preset target facial database, so as to obtain the target facial feature data that meets a preset similarity with the couple matching feature data in the target facial database, and output the opposite sex information corresponding to the target facial feature data. The target facial database is used to store the opposite sex facial data to be matched. The matchmaking and dating matching terminal 10 is used to obtain the gender information and facial image of the user to be matched. The blind date matching terminal 10 is also used to obtain facial feature data based on the facial image of the user to be matched, and send the gender information and facial feature data of the user to be matched to the blind date matching server 20. The blind date matching terminal 10 is also used to obtain the opposite sex information corresponding to the target facial feature data output by the blind date matching server 20, and display it. In one embodiment, the blind date matching terminal 10 includes a smart mobile terminal. In one embodiment, the blind date matching terminal 10 is a mobile phone, and the blind date and friendship matching method is implemented through the APP software on the mobile phone.

[0085] In the embodiment of the present application, a method for matching blind dates and making friends based on a deep learning neural network is disclosed. First, the gender information and facial image of the user to be matched are obtained, and facial feature data is obtained based on the facial image of the user to be matched; then the gender information and facial feature data of the user to be matched are input into a mathematical model for matching couples to obtain matching feature data of the couples, and the matching feature data of the couples are matched with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the matching feature data of the couples; finally, the opposite sex information corresponding to the target facial feature data is output. Since the gender information and facial image of the user to be matched are input into the mathematical model for matching couples to obtain matching feature data of the couples, and matching is performed based on the matching feature data of the couples, the satisfaction of the two parties in the first meeting of blind dates and making friends is improved.

[0086] Those skilled in the art will appreciate that all or part of the functions of the various methods in the above-mentioned embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, and the storage medium can include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to implement the above-mentioned functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above-mentioned functions can be implemented. In addition, when all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can also be stored in a storage medium such as a server, another computer, disk, optical disk, flash disk or mobile hard disk, and can be downloaded or copied and saved in the memory of the local device, or the system of the local device is updated, and when the program in the memory is executed by the processor, all or part of the functions in the above-mentioned embodiments can be implemented.

[0087] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A dating and friendship matching method based on deep learning neural network, characterized in that: include: Obtain the gender information and facial image of the user to be matched; Acquiring facial feature data based on the facial image of the user to be matched; Inputting the gender information and facial feature data of the user to be matched into a couple matching mathematical model, and obtaining the couple matching feature data output by the couple matching mathematical model; the couple matching feature data is conditional data that meets the couple appearance of the user to be matched; Matching the matching feature data of the couple with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the matching feature data of the couple, and outputting opposite-sex information corresponding to the target facial feature data; The target facial database is used to store facial data of the opposite sex to be matched; The method for obtaining the mathematical model for matching a couple comprises: Acquire a preset number of facial images of a couple whose marriage age meets a preset time period; Extracting facial feature data of the husband and wife respectively based on the facial images of the husband and wife; The facial feature data of the couple are used as elements in a training set to train the couple matching mathematical model to obtain the corresponding relationship between the facial feature data and the couple matching feature data.

2. The dating and making friends matching method according to claim 1, characterized in that: The step of acquiring facial feature data based on the facial image of the user to be matched includes: Performing face detection on the facial image and acquiring a facial image; Preprocessing the facial image; the preprocessing includes scale normalization and / or grayscale value correction; Acquire positioning feature points of the preprocessed face image; the positioning feature points are used to locate the face image, and the positioning feature points include eyes, ears, nose, chin, corners of mouth and / or center points of mouth; The facial feature data is acquired according to the positioning feature points.

3. The dating and friendship matching method according to claim 2, characterized in that: The facial feature data includes face shape feature data, eye shape feature data, nose shape feature data and / or mouth shape feature data; The facial feature data include face length, face width, forehead width, forehead height, forehead-zygomatic height difference, jaw width, jaw height, left frontal angle and / or right frontal angle; The eye shape feature data include the width of the outer eyelids of both eyes, the width of the inner eyelids of both eyes, the width of the inner and outer palpebral fissures of the left eye, the width of the left palpebral fissure, the height difference between the inner and outer eyes of the left eye, the outer palpebral fissure angle of the left eye, the inner palpebral fissure angle of the left eye, the width of the inner and outer palpebral fissure of the right eye, the width of the right palpebral fissure, the height difference between the inner and outer eyes of the right eye, the outer palpebral fissure angle of the right eye, the inner palpebral fissure angle of the right eye, the roundness of the left eye and / or the three-angle of the right eye, wherein the roundness of the left eye includes the polygonal area of ​​the left eye contour and / or the area of ​​the circumscribed circle of the left eye, and the three-angle of the right eye includes the area of ​​the inscribed triangle of the right eye and / or the polygonal area of ​​the right eye contour; The nose shape feature data includes nose length, nose width, height difference between nose tip and nose base, right nose wing width, height difference between left nostril and nose base, left nostril width and / or left nostril height; The mouth shape feature data include lip width, lip height, width from the midpoint of the upper lip to the left lip peak, width from the midpoint of the upper lip to the right lip peak, height from the midpoint of the upper lip to the left lip peak, height from the midpoint of the upper lip to the right lip peak, height from the left corner of the mouth to the left lip peak, height from the left corner of the mouth to the lip bead, height from the right corner of the mouth to the lip bead, height from the middle of the upper lip to the lip bead and / or height from the lip bead to the midpoint of the lower lip.

4. The dating and making friends matching method as claimed in claim 2, characterized in that: The facial feature data includes face shape, eye shape, nose shape and / or mouth shape; The eye shapes include droopy eyes, phoenix eyes, hanging eyes, near-centered eyes, squinting eyes, triangular eyes, elongated eyes, almond eyes, round eyes and / or telecentric eyes; The nose shapes include upturned nose, short flat nose, Nubian nose, Greek nose and / or hooked nose; The mouth shapes include the boat-shaped mouth, the bow-shaped mouth, the tiger's mouth, the four-character mouth, the upturned moon mouth and / or the cherry mouth; The face shapes include round, square, heart, oblong, diamond, oval and / or pear.

5. The dating and making friends matching method according to claim 1, characterized in that: Matching the matching feature data of the couple with a preset target face database includes: The opposite-sex facial data to be matched includes the opposite-sex facial image to be matched; The LBP operator is used to compare the matching feature data of the couple with the face images of the opposite sex in the target face database.

6. The dating and friendship matching method according to claim 1, characterized in that: Matching the matching feature data of the couple with a preset target face database includes: The opposite-sex facial data to be matched includes facial feature data of the opposite-sex to be matched; The matching feature data of the couple is matched with the facial feature data of the opposite sex to be matched.

7. A computer-readable storage medium, characterized in that: The method comprises a program which can be executed by a processor to implement the dating and friendship matching method as described in any one of claims 1 to 6.

8. A dating and friendship matching system based on deep learning neural network, characterized in that: It includes a blind date matching server, which is used to obtain gender information and facial feature data of a user to be matched; the facial feature data is obtained based on the facial image of the user to be matched; The matchmaking server is further used to input the gender information and facial feature data of the to-be-matched user into a couple matching mathematical model, and obtain the couple matching feature data output by the couple matching mathematical model; the couple matching feature data is conditional data that meets the couple appearance of the to-be-matched user; The matchmaking matching server is further used to match the matching feature data of the couple with a preset target facial database to obtain target facial feature data in the target facial database that meets a preset similarity with the matching feature data of the couple, and output opposite-sex information corresponding to the target facial feature data; The target facial database is used to store facial data of the opposite sex to be matched; The method for obtaining the mathematical model for matching a couple comprises: Acquire a preset number of facial images of a couple whose marriage age meets a preset time period; Extracting facial feature data of the husband and wife respectively based on the facial images of the husband and wife; The facial feature data of the couple are used as elements in a training set to train the couple matching mathematical model to obtain the corresponding relationship between the facial feature data and the couple matching feature data.

9. The dating and friendship matching system according to claim 8, characterized in that: Also included is a blind date matching terminal for obtaining the gender information and facial image of the user to be matched; The blind date matching terminal is also used to obtain facial feature data based on the facial image of the to-be-matched user, and send the gender information and facial feature data of the to-be-matched user to the blind date matching server; And / or, the blind date matching terminal is further used to obtain the opposite-sex information corresponding to the target facial feature data output by the blind date matching server and display it.

Images