Identity recognition method and system based on face image processing

Abstract

The present invention discloses an identity recognition method and system based on face image processing, belonging to the technical field of image processing and recognition; by acquiring and processing the feature coordinates of different features in the face image, and combining and collecting and processing the center point coordinates of different features; by inputting a feature parameter analysis set into a face image information library to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence; in the case that the necessity analysis of different local feature recognition verification dimensions is passed in the early stage, the consistency analysis of the overall dimensional features is further performed on the feature detection identification sequence that meets the minimum verification dimension necessary conditions, thereby realizing cross-feature data verification and analysis, and improving the diversity and reliability of face image identity recognition; the present invention can solve the technical problem that the data processing and analysis dimension of face image identity recognition in the existing solution is single, resulting in poor identity recognition effect.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing and recognition, and in particular to an identity recognition method and system based on face image processing. Background Art

[0002] Facial image processing has a wide range of applications in the field of identity recognition, and identity recognition is mainly achieved through the following technologies: face detection, facial feature extraction, face matching, deep learning, liveness detection, and multimodal fusion. The comprehensive use of the above technologies can achieve accurate and fast facial image processing identity recognition.

[0003] However, when implementing existing identity recognition schemes based on facial image processing, after identifying different local features, the necessity of different local feature recognition and verification dimensions is not analyzed. When the necessity analysis of different local feature recognition and verification dimensions is passed, the consistency analysis of the overall dimensional features is further performed on the feature detection identification sequence that meets the minimum verification dimension requirements. Cross-feature data verification and analysis cannot be performed, and there is a defect that the data processing and analysis dimension of facial image identity recognition is single, resulting in poor identity recognition effect. Summary of the invention

[0004] The purpose of the present invention is to provide an identity recognition method and system based on face image processing, which is used to solve the technical problem that the existing solution of face image identity recognition has a single data processing and analysis dimension, resulting in poor identity recognition effect.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The identity recognition method based on face image processing includes:

[0007] Acquire a face image and perform image preprocessing and feature extraction to obtain a feature image;

[0008] Acquire a first feature, a second feature, a third feature, and a fourth feature in the feature image and combine them to obtain a feature set;

[0009] Obtain the midpoint of the feature image and set it as the coordinate origin, and construct an image coordinate system according to the preset coordinate spacing and coordinate direction, obtain the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates corresponding to the first feature, the second feature and the third feature in the feature set according to the image coordinate system, and combine them to obtain a feature coordinate set;

[0010] The coordinate parameters of different feature contours in the feature set are identified and processed to obtain a feature contour processing set; the coordinate parameters of different feature distances in the feature coordinate set are processed to obtain a feature distance processing set;

[0011] Input the feature parameter analysis set into the face image information database to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence; the feature detection identification sequence is composed of a number of elements with values ​​of 0 or 1;

[0012] A verification dimension necessity analysis is performed on the feature detection identification sequence, and based on the analysis results, feature distance screening and identification is performed on the feature distance array and the identity recognition result is output.

[0013] Preferably, two vertices at the horizontal positions corresponding to the first feature and the second feature in the feature set are obtained, and the coordinates of all contour coordinate points of the lower feature contour between the two vertices of the first feature and the second feature are respectively obtained and combined to obtain a first contour coordinate set and a second contour coordinate set;

[0014] Obtaining and combining the coordinates of all contour coordinate points of the feature contour corresponding to the third feature to obtain a third contour coordinate set;

[0015] Respectively obtain all contour coordinate point coordinates of the upper feature contour and the lower feature contour between two vertices of the fourth feature and combine them to obtain a fourth contour coordinate set and a fifth contour coordinate set;

[0016] The first contour coordinate set, the second contour coordinate set, the third contour coordinate set, the fourth contour coordinate set and the fifth contour coordinate set are arranged and combined in sequence to obtain a feature coordinate set.

[0017] Preferably, the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates in the feature coordinate set are obtained;

[0018] According to the coordinate distance calculation formula, respectively calculate the first distance value between the first feature center point coordinates and the second feature center point coordinates, the second distance value between the first feature center point coordinates and the third feature center point coordinates, the third distance value between the first feature center point coordinates and the fourth feature center point coordinates, and the fourth distance value between the third feature center point coordinates and the fourth feature center point coordinates;

[0019] The values ​​of the first distance value, the second distance value, the third distance value and the fourth distance value are extracted and sorted and combined to obtain a feature distance array.

[0020] Preferably, the feature coordinate set in the feature parameter analysis set is input into a face image information library, and a feature detection identification sequence is output through a feature coordinate detection model.

[0021] Preferably, the feature coordinate detection model J({(xi, yi)}) is expressed as ; In the formula: {(xi, yi)} are different feature coordinate sets; (xi, yi) are different contour coordinate point coordinates in the feature coordinate set, i is a different contour coordinate point in the feature coordinate set, i=1, 2, 3, ..., n; n is a positive integer, representing the total number of contour coordinate points in the corresponding feature coordinate set.

[0022] Preferably, when there is sample data in the face image information library that is completely identical to the feature coordinate set in the feature parameter analysis set, a feature detection identifier with a value of 1 is output; otherwise, a feature detection identifier with a value of 0 is output;

[0023] According to the feature detection identifier with a value of 1, the sample person images associated with the sample data of the same feature coordinate set as the feature parameter analysis set in the person image information library are marked as preliminary sample images, and all the preliminary sample images belonging to the same feature coordinate set are combined to obtain a preliminary sample image set and associated with the corresponding feature detection identifier;

[0024] The numerical values ​​of the feature detection identifiers corresponding to the output of all feature coordinate sets in the feature parameter analysis set are extracted and combined to obtain a feature detection identifier sequence.

[0025] Preferably, when performing verification dimension necessity analysis according to the feature detection identification sequence, the feature detection identification sequence is traversed to count the total number of elements with a value of 1 and marked as a standard verification necessary value;

[0026] If the standard verification necessary value is less than N, where N is a positive integer, then an invalid verification dimension instruction is generated, and a prompt indicating that the identity corresponding to the generated face image does not exist is generated;

[0027] Otherwise, a verification dimension valid instruction is generated, and according to the verification dimension valid instruction, the feature distance array is used to implement feature distance screening and recognition on the preliminary sample image set.

[0028] Preferably, a preliminary sample image set associated with all elements with a value of 1 in the feature detection identification sequence and a corresponding sample feature spacing array are obtained, and the feature spacing array in the feature parameter analysis set and all sample feature spacing arrays are subjected to consistency analysis through a feature spacing detection model and an analysis identification is output; the analysis identification includes a value of 0 or 1;

[0029] According to the analysis flag with a value of 1, the portrait identity number associated with the corresponding sample feature spacing array is analyzed. If the portrait identity number appears in all the preliminary selected sample image sets, the identity corresponding to the portrait identity number is marked as the first identification identity and a prompt is output;

[0030] If the total number of times the portrait identity number appears in all the preliminary sample image sets is greater than or equal to N, the identity corresponding to the portrait identity number is marked as a suspected identification identity and a prompt is output.

[0031] Preferably, the characteristic spacing detection model Z(k) is expressed as ; In the formula: k is the different element values ​​in the feature spacing array, k=1, 2, 3, 4; j represents different sample feature spacing arrays, jk0 is the different element values ​​in different sample feature spacing arrays.

[0032] The identity recognition system based on face image processing includes:

[0033] The face image processing module is used to obtain the face image and perform image preprocessing and feature extraction to obtain a feature image; obtain the first feature, the second feature, the third feature and the fourth feature in the feature image and combine them to obtain a feature set;

[0034] An image feature coordinate acquisition module is used to acquire the midpoint of the feature image and set it as the coordinate origin, and to construct an image coordinate system according to a preset coordinate spacing and coordinate direction, and to acquire the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates corresponding to the first feature, the second feature and the third feature in the feature set according to the image coordinate system and combine them to obtain a feature coordinate set;

[0035] The image feature coordinate processing module is used to identify and process the coordinate parameters of different feature contours in the feature set to obtain a feature contour processing set; and to process the coordinate parameters of different feature distances in the feature coordinate set to obtain a feature distance processing set;

[0036] The image feature coordinate analysis module is used to input the feature parameter analysis set into the face image information library to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence;

[0037] The face image identity verification module is used to perform verification dimension necessity analysis on the feature detection identification sequence, and to perform feature distance screening and identification on the feature distance array based on the analysis results and output the identity recognition result.

[0038] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0039] The present invention acquires and processes the feature coordinates of different features in the facial image and combines them, as well as collects and processes the center point coordinates of different features, so as to provide cross-feature connection data support for the subsequent recognition and verification of facial images with appearance changes.

[0040] The present invention obtains a feature detection identification sequence by inputting a feature parameter analysis set into a face image information library for feature parameter retrieval and comparison. It can then perform modular recognition and analysis on face images from different local feature dimensions, thereby obtaining the recognition and matching status of local features at different positions in the face image and providing reliable local data support for subsequent recognition and analysis of the overall dimension of the face image.

[0041] When the necessity analysis of different local feature recognition and verification dimensions is passed in the early stage, the present invention further implements consistency analysis of the overall dimensional features on the feature detection identification sequence that meets the necessary conditions of the minimum verification dimension, realizes cross-feature data verification and analysis, and improves the diversity and reliability of facial image identity recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention will be further described below in conjunction with the accompanying drawings.

[0043] Figure 1 The figure is a flowchart of the identity recognition method based on face image processing of the present invention.

[0044] Figure 2 It is a flowchart of implementing the verification dimension necessity analysis according to the feature detection identification sequence in the present invention.

[0045] Figure 3 The module block diagram of the identity recognition system based on face image processing of the present invention. DETAILED DESCRIPTION

[0046] 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 operation and maintenance personnel in this field without creative work are within the scope of protection of the present invention.

[0047] Example 1: Figure 1 As shown, the present invention is an identity recognition method based on face image processing, comprising:

[0048] Acquire a face image and perform image preprocessing and feature extraction to obtain a feature image;

[0049] The image preprocessing includes but is not limited to denoising, smoothing, sharpening, contrast enhancement and cropping, and the feature extraction is an existing conventional image feature extraction algorithm. The specific steps are not described here.

[0050] Acquire a first feature, a second feature, a third feature, and a fourth feature in the feature image and combine them to obtain a feature set;

[0051] Specifically, the feature image is a facial image including several features; the first feature, the second feature, the third feature and the fourth feature are the left eye, the right eye, the nose and the lips respectively; the specific contour ranges of the left eye, the right eye, the nose and the lips can be determined based on the existing recognition design scheme;

[0052] Obtain the midpoint of the feature image and set it as the coordinate origin, and construct an image coordinate system according to the preset coordinate spacing and coordinate direction, obtain the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates corresponding to the first feature, the second feature and the third feature in the feature set according to the image coordinate system, and combine them to obtain a feature coordinate set;

[0053] The preset coordinate spacing and coordinate direction are determined according to the sample design parameters corresponding to all facial feature coordinate data stored in the facial image information library, so as to ensure that the consistency analysis of the sample feature coordinates can be implemented later;

[0054] The coordinate parameters of different feature contours in the feature set are identified and processed to obtain a feature contour processing set; including:

[0055] Obtain two vertices at the horizontal positions corresponding to the first feature and the second feature in the feature set, specifically the intersection points on both sides of the upper and lower eyelids, respectively obtain the coordinates of all contour points of the lower feature contour between the two vertices of the first feature and the second feature, and combine them to obtain the first contour coordinate set and the second contour coordinate set;

[0056] Obtaining and combining the coordinates of all contour coordinate points of the feature contour corresponding to the third feature to obtain a third contour coordinate set;

[0057] Respectively obtain all the contour coordinate points of the upper feature contour and the lower feature contour between the two vertices of the fourth feature and combine them to obtain a fourth contour coordinate set and a fifth contour coordinate set, specifically the intersection points on both sides of the upper and lower lips;

[0058] Arrange and combine the first contour coordinate set, the second contour coordinate set, the third contour coordinate set, the fourth contour coordinate set and the fifth contour coordinate set in order to obtain a feature coordinate set;

[0059] In the embodiment of the present invention, by acquiring and processing the feature coordinates of different features in the face image, reliable data support can be provided for subsequent recognition and analysis of different local features of the face image.

[0060] Processing the coordinate parameters of different feature distances in the feature coordinate set to obtain a feature distance processing set; including:

[0061] Obtain the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates, and the fourth feature center point coordinates in the feature coordinate set;

[0062] According to the coordinate distance calculation formula, respectively calculate the first distance value between the first feature center point coordinates and the second feature center point coordinates, the second distance value between the first feature center point coordinates and the third feature center point coordinates, the third distance value between the first feature center point coordinates and the fourth feature center point coordinates, and the fourth distance value between the third feature center point coordinates and the fourth feature center point coordinates;

[0063] Extract the values ​​of the first distance value, the second distance value, the third distance value and the fourth distance value and sort and combine them to obtain a feature distance array;

[0064] It should be noted that, due to the time difference between the collected face image and the sample face image collected and stored in advance, some features of the actual face image will change. For example, the face image of a person who has gained weight will change, but the coordinates of the center points corresponding to each feature will not change significantly. Therefore, in the embodiment of the present invention, by collecting and processing the coordinates of the center points of different features, cross-feature association data support can be provided for subsequent face image recognition verification of appearance changes;

[0065] Input the feature parameter analysis set into the face image information database to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence; including:

[0066] Input the feature coordinate set in the feature parameter analysis set into the face image information database, and output the feature detection identification sequence through the feature coordinate detection model; the feature detection identification sequence is composed of a number of elements with values ​​of 0 or 1; the face image information database can be called by the existing database after authorization;

[0067] Among them, the feature coordinate detection model J({(xi, yi)}) is expressed as ; Where: {(xi, yi)} is a different feature coordinate set; (xi, yi) is a different contour coordinate point in the feature coordinate set, i is a different contour coordinate point in the feature coordinate set, i=1, 2, 3, ..., n; n is a positive integer, representing the total number of contour coordinate points in the corresponding feature coordinate set;

[0068] In addition, when there is sample data in the face image information database that is completely identical to the feature coordinate set in the feature parameter analysis set, a feature detection flag with a value of 1 is output;

[0069] Otherwise, the feature detection flag with a value of 0 is output;

[0070] According to the feature detection identifier with a value of 1, the sample person images associated with the sample data of the same feature coordinate set as the feature parameter analysis set in the person image information library are marked as preliminary sample images, and all the preliminary sample images belonging to the same feature coordinate set are combined to obtain a preliminary sample image set and associated with the corresponding feature detection identifier;

[0071] Extracting the values ​​of the feature detection identifiers corresponding to the output of all feature coordinate sets in the feature parameter analysis set and combining them to obtain a feature detection identifier sequence;

[0072] In the embodiment of the present invention, by inputting the feature parameter analysis set into the face image information library to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence, the face image can be modularly recognized and analyzed from different local feature dimensions, which can not only obtain the recognition and matching status of local features at different positions in the face image, but also provide reliable local data support for the subsequent recognition analysis of the overall dimension of the face image;

[0073] Perform verification dimension necessity analysis on the feature detection identification sequence, and perform feature distance screening and identification on the feature distance array based on the analysis results and output the identity recognition result; including:

[0074] like Figure 2 As shown, when implementing the verification dimension necessity analysis according to the feature detection identification sequence, the feature detection identification sequence is traversed to count the total number of elements with a value of 1 and marked as a necessary value for standard verification;

[0075] If the standard verification necessary value is less than N, where N is a positive integer and the value of N is 2, for example, only the feature of two eyes is the minimum necessary condition for verification dimension, then an invalid verification dimension instruction is generated, and a prompt indicating that the identity corresponding to the generated face image does not exist is generated;

[0076] Otherwise, a verification dimension effective instruction is generated, and according to the verification dimension effective instruction, the feature distance array is used to implement feature distance screening and recognition on the preliminary sample image set;

[0077] It should be noted that when the total number of local features of the recognition analysis does not meet the requirements, the subsequent overall dimension verification analysis will not be performed, which can effectively save the use of recognition resources and reduce the operating pressure of recognition hardware resources;

[0078] Among them, the preliminary sample image set associated with all elements with a value of 1 in the feature detection identification sequence and the corresponding sample feature spacing array are obtained, and the feature spacing array in the feature parameter analysis set and all sample feature spacing arrays are subjected to consistency analysis through the feature spacing detection model and the analysis identification is output;

[0079] Among them, the expression of the feature spacing detection model Z(k) is: ; Where: k is the value of different elements in the feature spacing array, k=1, 2, 3, 4; j represents different sample feature spacing arrays, jk0 is the value of different elements in different sample feature spacing arrays;

[0080] According to the analysis flag with a value of 1, the portrait identity number associated with the corresponding sample feature spacing array is analyzed. If the portrait identity number appears in all the preliminary selected sample image sets, the identity corresponding to the portrait identity number is marked as the first identification identity and a prompt is output;

[0081] If the total number of times the portrait identity number appears in all the preliminary sample image sets is greater than or equal to N, the identity corresponding to the portrait identity number is marked as a suspected identification identity and a prompt is output.

[0082] In an embodiment of the present invention, when the necessity analysis of different local feature recognition and verification dimensions is passed in the early stage, the consistency analysis of the overall dimensional features is further implemented on the feature detection identification sequence that meets the necessary conditions of the minimum verification dimension, thereby realizing cross-feature data verification and analysis, and improving the diversity and reliability of facial image identity recognition.

[0083] Example 2: Figure 3 As shown, the identity recognition system based on face image processing includes:

[0084] The face image processing module is used to obtain the face image and perform image preprocessing and feature extraction to obtain a feature image; obtain the first feature, the second feature, the third feature and the fourth feature in the feature image and combine them to obtain a feature set;

[0085] An image feature coordinate acquisition module is used to acquire the midpoint of the feature image and set it as the coordinate origin, and to construct an image coordinate system according to a preset coordinate spacing and coordinate direction, and to acquire the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates corresponding to the first feature, the second feature and the third feature in the feature set according to the image coordinate system and combine them to obtain a feature coordinate set;

[0086] The image feature coordinate processing module is used to identify and process the coordinate parameters of different feature contours in the feature set to obtain a feature contour processing set; and to process the coordinate parameters of different feature distances in the feature coordinate set to obtain a feature distance processing set;

[0087] The image feature coordinate analysis module is used to input the feature parameter analysis set into the face image information library to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence;

[0088] The face image identity verification module is used to perform verification dimension necessity analysis on the feature detection identification sequence, and perform feature distance screening and identification on the feature distance array based on the analysis results and output the identity recognition result.

[0089] In addition, the formulas involved in the above are all dimensionless and numerical calculations. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and simulating it with simulation software.

[0090] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.

[0091] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0093] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the basic features of the present invention.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, ordinary operation and maintenance personnel in the field should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An identity recognition method based on face image processing, characterized in that: include: Acquire a face image and perform image preprocessing and feature extraction to obtain a feature image; Acquire a first feature, a second feature, a third feature, and a fourth feature in the feature image and combine them to obtain a feature set; The first feature, the second feature, the third feature and the fourth feature are respectively the left eye, the right eye, the nose and the lips; Obtain the midpoint of the feature image and set it as the coordinate origin, and construct an image coordinate system according to the preset coordinate spacing and coordinate direction, obtain the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates corresponding to the first feature, the second feature and the third feature in the feature set according to the image coordinate system and combine them, calculate the distance values ​​between the center point coordinates of different features according to the coordinate distance calculation formula and combine them to obtain a feature distance array; Identify and process the coordinate parameters of different feature contours in the feature set to obtain a feature coordinate set; Input all feature coordinate sets in the feature parameter analysis set into the face image information database to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence; the feature detection identification sequence is composed of a number of elements with values ​​of 0 or 1; Perform verification dimension necessity analysis on the feature detection identification sequence, and perform feature distance screening and identification on the feature distance array based on the analysis results and output the identity recognition result; When performing verification dimension necessity analysis based on the feature detection identification sequence, the feature detection identification sequence is traversed to count the total number of elements with a value of 1 and marked as a necessary value for standard verification; If the standard verification necessary value is less than N, where N is a positive integer, then an invalid verification dimension instruction is generated, and a prompt indicating that the identity corresponding to the generated face image does not exist is generated; Otherwise, a verification dimension valid instruction is generated, and according to the verification dimension valid instruction, the feature distance array is used to implement feature distance screening and recognition on the preliminary sample image set.

2. The identity recognition method based on face image processing according to claim 1, characterized in that: Obtain two vertices at horizontal positions corresponding to the first feature and the second feature in the feature set, respectively obtain all contour coordinate point coordinates of the lower feature contour between the two vertices of the first feature and the second feature, and combine them to obtain a first contour coordinate set and a second contour coordinate set; Obtaining and combining the coordinates of all contour coordinate points of the feature contour corresponding to the third feature to obtain a third contour coordinate set; Respectively obtain all contour coordinate point coordinates of the upper feature contour and the lower feature contour between two vertices of the fourth feature and combine them to obtain a fourth contour coordinate set and a fifth contour coordinate set; The first contour coordinate set, the second contour coordinate set, the third contour coordinate set, the fourth contour coordinate set and the fifth contour coordinate set are arranged and combined in sequence to obtain a feature coordinate set.

3. The identity recognition method based on face image processing according to claim 2 is characterized in that: Obtain the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates, and the fourth feature center point coordinates in the feature coordinate set; According to the coordinate distance calculation formula, respectively calculate the first distance value between the first feature center point coordinates and the second feature center point coordinates, the second distance value between the first feature center point coordinates and the third feature center point coordinates, the third distance value between the first feature center point coordinates and the fourth feature center point coordinates, and the fourth distance value between the third feature center point coordinates and the fourth feature center point coordinates; The values ​​of the first distance value, the second distance value, the third distance value and the fourth distance value are extracted and sorted and combined to obtain a feature distance array.

4. The identity recognition method based on face image processing according to claim 2 is characterized in that: The feature coordinate set in the feature parameter analysis set is input into the face image information library, and the feature detection identification sequence is output through the feature coordinate detection model.

5. The identity recognition method based on face image processing according to claim 4 is characterized in that: The feature coordinate detection model J({(xi, yi)}) is expressed as ; In the formula: {(xi, yi)} are different feature coordinate sets; (xi, yi) are different contour coordinate point coordinates in the feature coordinate set, i is a different contour coordinate point in the feature coordinate set, i=1, 2, 3, ..., n; n is a positive integer, representing the total number of contour coordinate points in the corresponding feature coordinate set.

6. The identity recognition method based on face image processing according to claim 4 is characterized in that: When there is sample data in the face image information database that is completely identical to the feature coordinate set in the feature parameter analysis set, a feature detection flag with a value of 1 is output; otherwise, a feature detection flag with a value of 0 is output; According to the feature detection identifier with a value of 1, the sample person images associated with the sample data of the same feature coordinate set as the feature parameter analysis set in the person image information library are marked as preliminary sample images, and all the preliminary sample images belonging to the same feature coordinate set are combined to obtain a preliminary sample image set and associated with the corresponding feature detection identifier; The numerical values ​​of the feature detection identifiers corresponding to the output of all feature coordinate sets in the feature parameter analysis set are extracted and combined to obtain a feature detection identifier sequence.

7. The identity recognition method based on face image processing according to claim 1, characterized in that: Obtain the preliminary sample image set associated with all elements with a value of 1 in the feature detection identification sequence and the corresponding sample feature spacing array, and perform consistency analysis on the feature spacing array in the feature parameter analysis set and all sample feature spacing arrays through the feature spacing detection model and output the analysis identification; the analysis identification contains a value of 0 or 1; According to the analysis flag with a value of 1, the portrait identity number associated with the corresponding sample feature spacing array is analyzed. If the portrait identity number appears in all the preliminary selected sample image sets, the identity corresponding to the portrait identity number is marked as the first identification identity and a prompt is output; If the total number of times the portrait identity number appears in all the preliminary sample image sets is greater than or equal to N, the identity corresponding to the portrait identity number is marked as a suspected identification identity and a prompt is output.

8. The identity recognition method based on face image processing according to claim 7 is characterized in that the feature distance detection model Z(k) is expressed as ; In the formula: k is the different element values ​​in the feature spacing array, k=1, 2, 3, 4; j represents different sample feature spacing arrays, jk0 is the different element values ​​in different sample feature spacing arrays.

9. An identity recognition system based on face image processing, applied to the identity recognition method based on face image processing as claimed in any one of claims 1 to 8, characterized in that: include: The face image processing module is used to obtain the face image and perform image preprocessing and feature extraction to obtain a feature image; Acquire a first feature, a second feature, a third feature, and a fourth feature in the feature image and combine them to obtain a feature set; An image feature coordinate acquisition module is used to acquire the midpoint of the feature image and set it as the coordinate origin, and to construct an image coordinate system according to a preset coordinate spacing and coordinate direction, and to acquire the first feature center point coordinates, the second feature center point coordinates, the third feature center point coordinates and the fourth feature center point coordinates corresponding to the first feature, the second feature and the third feature in the feature set according to the image coordinate system and combine them to obtain a feature coordinate set; The image feature coordinate processing module is used to identify and process the coordinate parameters of different feature contours in the feature set to obtain a feature coordinate set; and process the coordinate parameters of different feature distances in the feature coordinate set to obtain a feature distance array; The image feature coordinate analysis module is used to input the feature parameter analysis set into the face image information library to perform feature parameter retrieval and comparison to obtain a feature detection identification sequence; The face image identity verification module is used to perform verification dimension necessity analysis on the feature detection identification sequence, and perform feature distance screening and identification on the feature distance array based on the analysis results and output the identity recognition result.

Images