Personnel identification method and device, electronic equipment and storage medium

Abstract

The invention discloses a personnel identification method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining a face image of a target object and a feature matrix of a comparison object; performing face region extraction on the face image to obtain a face key part of the target object; performing feature extraction on the face image and the face key parts to obtain a feature point matrix; according to the feature point matrix and the feature matrix, sequentially processing based on distance similarity to obtain a similarity score of each target area of the target object and the comparison object; and carrying out integration operation on the similarity scores of all the target areas to obtain target identification scores of the target object and the comparison object. According to the method, through a multi-layer and multi-region feature extraction and similarity calculation method, the recognition accuracy and robustness of the target object and the comparison object are remarkably improved, meanwhile, the recognition efficiency is improved, and the method has good flexibility and expandability and can be widely applied to the technical field of data processing.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a personnel identification method, device, electronic equipment and storage medium. Background Art

[0002] In current scenarios such as security monitoring, crowd management, and identity authentication, accurate identification of target detection personnel is a crucial link. However, existing technical methods face many challenges and shortcomings when dealing with this task.

[0003] Traditionally, the identification of target detection personnel mainly relies on manual verification. Although this method can ensure the accuracy of identification to a certain extent, its significant disadvantages are large workload and low efficiency. With the popularization of equipment such as surveillance cameras, the amount of video and image data generated has increased dramatically. Relying solely on manual verification is not only time-consuming and laborious, but also prone to misjudgment or omission due to human factors (such as fatigue, distraction, etc.), thereby reducing the reliability and practicality of the overall identification.

[0004] In order to overcome the limitations of manual verification, some technologies have begun to try to use computer vision and image processing technology to achieve automatic identification of target detection personnel. However, these technologies often rely solely on overall similarity for matching, that is, by calculating the overall similarity between the image of the person to be identified and the images already stored in the database to determine whether they are the same person. Although this method has improved the recognition efficiency to a certain extent, its recognition technology is relatively simple and lacks detailed analysis and processing of the differences in characteristics of different individuals.

[0005] In addition, this method is also prone to mismatching for different individuals with similar appearance features, thereby reducing the accuracy and robustness of recognition.

[0006] Therefore, the existing technologies have obvious deficiencies and limitations in identifying target detection personnel, whether it is manual verification or automatic recognition methods that simply rely on overall similarity. Summary of the invention

[0007] The present invention proposes a personnel identification method, device, electronic device and storage medium, aiming to solve one of the technical problems in the related art at least to a certain extent. The embodiments of the present invention can perform personnel identification efficiently and accurately.

[0008] In one aspect, an embodiment of the present invention provides a method for identifying a person, comprising:

[0009] Obtaining a face image of a target object and a feature matrix of a comparison object;

[0010] Extract the face area of ​​the face image to obtain the key face parts of the target object;

[0011] Extract features from face images and key face parts to obtain feature point matrix;

[0012] The feature point matrix includes a first feature point matrix of the face area of ​​the target object and a second feature point matrix of the specific area of ​​each key part of the face; the feature matrix includes a first feature matrix of the face area of ​​the comparison object and a second feature matrix of the specific area;

[0013] According to the feature point matrix and the feature matrix, the similarity score of each target area of ​​the target object and the comparison object is obtained by sequentially processing based on the distance similarity; the target area includes the face area and each specific area;

[0014] The similarity scores of all target regions are integrated to obtain the target recognition scores of the target object and the comparison object.

[0015] Optionally, obtaining a face image of a target object includes the following steps:

[0016] Capturing a first image of the target object through a video device; the first image contains a face area of ​​the target object;

[0017] The face area of ​​the first image is located and recognized, and then a face image is obtained through a cropping operation.

[0018] Optionally, the feature matrix of the comparison object is obtained from a preset multi-dimensional feature base library, and the method further includes the following steps:

[0019] Obtain image data of all target persons and set a unique ID for each target person in the preset management library;

[0020] Based on the image data, the overall face region feature extraction and key part region feature extraction are performed in sequence to obtain the feature matrix of each target person;

[0021] The feature matrix corresponding to each target person is associated with a unique identifier and stored in different regions to construct a multi-dimensional feature base.

[0022] The multi-dimensional feature base includes a first feature base of the overall face area of ​​all target persons and a second feature base of each key part area.

[0023] Optionally, performing face region extraction on the face image to obtain key face parts of the target object includes the following steps:

[0024] Perform facial feature point detection on the face image to obtain the key facial feature points of the target object;

[0025] Based on the key feature points of the face, the face image is cropped to obtain the key parts of the face of the target object;

[0026] Among them, the key feature points of the face represent the regional positions of the key parts of the face; the key parts of the face include the eyes, nose and mouth.

[0027] Optionally, feature extraction is performed on the face image and key face parts to obtain a feature point matrix, including the following steps:

[0028] By using a preset feature extraction algorithm, a preset number of feature points are extracted from the face image and the target area of ​​each key part of the face in turn;

[0029] Based on all the feature points of each target area, a first feature point matrix of the face area and a second feature point matrix of a specific area of ​​each key part of the face are obtained.

[0030] Optionally, according to the feature point matrix and the feature matrix, the similarity scores of each target area of ​​the target object and the comparison object are obtained by sequentially processing based on the distance similarity, including the following steps:

[0031] According to the feature point matrix and the feature matrix, the feature matrix feature difference corresponding to each target area is calculated in turn;

[0032] The feature differences of all feature matrices are normalized to obtain the similarity score of each target area between the target object and the comparison object.

[0033] Optionally, integrating the similarity scores of all target regions to obtain target recognition scores of the target object and the comparison object includes the following steps:

[0034] Based on the preset weight value of each target area, the similarity scores of all target areas are weighted and summed to obtain the target recognition score of the target object and the comparison object.

[0035] On the other hand, an embodiment of the present invention provides a person identification device, including:

[0036] The first module is used to obtain the face image of the target object and the feature matrix of the comparison object;

[0037] The second module is used to extract the face area of ​​the face image to obtain the key face parts of the target object;

[0038] The third module is used to extract features of face images and key face parts to obtain a feature point matrix;

[0039] The feature point matrix includes a first feature point matrix of the face area of ​​the target object and a second feature point matrix of the specific area of ​​each key part of the face; the feature matrix includes a first feature matrix of the face area of ​​the comparison object and a second feature matrix of the specific area;

[0040] The fourth module is used to process the feature point matrix and the feature matrix in sequence based on the distance similarity to obtain the similarity score of each target area of ​​the target object and the comparison object; the target area includes the face area and each specific area;

[0041] The fifth module is used to integrate the similarity scores of all target areas to obtain the target recognition scores of the target object and the comparison object.

[0042] Optionally, the feature matrix of the comparison object is obtained from a preset multi-dimensional feature base library, and the device further includes:

[0043] The sixth module is used to obtain the image data of all target persons and set a unique identification for each target person in the preset management library;

[0044] The seventh module is used to extract the overall face area features and key part area features in turn according to the image data to obtain the feature matrix of each target person;

[0045] The eighth module is used to associate the feature matrix corresponding to each target person with the unique identifier and store them in different regions to construct a multi-dimensional feature base library;

[0046] The multi-dimensional feature base includes a first feature base of the overall face area of ​​all target persons and a second feature base of each key part area.

[0047] On the other hand, an embodiment of the present invention provides an electronic device, including: a processor and a memory; the memory is used to store programs; the processor executes the program to implement the above-mentioned personnel identification method.

[0048] On the other hand, an embodiment of the present invention provides a computer storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the above-mentioned personnel identification method.

[0049] The embodiment of the present invention obtains a facial image of a target object and a feature matrix of a comparison object; extracts facial regions from the facial image to obtain key facial parts of the target object; extracts features from the facial image and key facial parts to obtain a feature point matrix; wherein the feature point matrix includes a first feature point matrix of the facial region of the target object and a second feature point matrix of specific regions of each key facial part; the feature matrix includes a first feature matrix of the facial region of the comparison object and a second feature matrix of specific regions; based on the feature point matrix and the feature matrix, the similarity score of each target region of the comparison object is obtained by sequentially processing based on distance similarity; the target region includes a facial region and each specific region; the similarity scores of all target regions are integrated to obtain the target recognition score of the target object and the comparison object. The embodiment of the present invention includes at least the following beneficial effects:

[0050] Improve recognition accuracy: The present invention extracts the facial area and key parts in the facial image of the target object, and performs feature extraction respectively to generate a feature point matrix including a first feature point matrix and a second feature point matrix. This multi-level, multi-region feature extraction method can more comprehensively capture the facial features of the target object, thereby improving recognition accuracy. At the same time, the feature matrix of the comparison object is also similarly divided and processed, so that in the subsequent similarity calculation, a more precise match can be performed for each target area, further reducing the possibility of mismatching.

[0051] Enhanced robustness of recognition: Since the present invention takes into account the specific areas of the face area and multiple key parts, even if the feature information of a certain area is disturbed or lost in some cases (such as lighting changes, partial occlusion, etc.), it can be supplemented and corrected by the feature information of other areas, thereby enhancing the robustness of recognition. In addition, by integrating the similarity scores of all target areas, the abnormal scores of individual areas can be further smoothed, improving the stability and reliability of overall recognition.

[0052] Improve recognition efficiency: The present invention can significantly reduce the amount of calculation and improve recognition efficiency by first extracting the face area and key parts, and then performing feature extraction and similarity calculation. This method has a more obvious advantage when performing person recognition on large-scale data sets.

[0053] At the same time, due to the use of a processing method based on distance similarity, the similarity score between the target object and the comparison object can be quickly calculated, thereby accelerating the recognition speed.

[0054] Flexibility and scalability: The present invention is not only applicable to the recognition of face areas, but can also be extended to the recognition of other specific areas according to actual needs. For example, in identity authentication scenarios, specific area recognition of eyes, nose, mouth and other parts can be added to improve the accuracy and security of recognition.

[0055] In addition, with the continuous development of technology, more advanced feature extraction methods and similarity calculation methods can be further introduced to continuously improve the performance and effect of recognition.

[0056] In summary, the present invention significantly improves the recognition accuracy and robustness of target objects and comparison objects through multi-level and multi-region feature extraction and similarity calculation methods, while improving recognition efficiency and having good flexibility and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0058] Figure 1 A schematic diagram of an implementation environment for personnel identification provided by an embodiment of the present invention;

[0059] Figure 2 A schematic diagram of a flow chart of a method for identifying a person provided by an embodiment of the present invention;

[0060] Figure 3 A schematic diagram of the expanded process of acquiring a facial image of a target object provided by an embodiment of the present invention;

[0061] Figure 4 A schematic diagram of an extended process of a personnel identification method provided in an embodiment of the present invention;

[0062] Figure 5 A schematic diagram of the expansion process of step S200 provided in an embodiment of the present invention;

[0063] Figure 6 A schematic diagram of the expansion process of step S300 provided in an embodiment of the present invention;

[0064] Figure 7 A schematic diagram of the expansion process of step S400 provided in an embodiment of the present invention;

[0065] Figure 8 A schematic diagram of an overall process example of a person identification method provided by an embodiment of the present invention;

[0066] Fig. 9 A schematic diagram of the structure of a personnel identification device provided by an embodiment of the present invention;

[0067] Fig.10 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] It should be noted that, although the functional modules are divided in the system schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first / S100", "second / S200", etc. in the specification, claims and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0070] Reference to "embodiments" in the present invention means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in the present invention may be combined with other embodiments.

[0071] like Figure 1 FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present invention. Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected to a network wirelessly or wired to complete data transmission and exchange.

[0072] Server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0073] In addition, the server 101 can also be a node server in the blockchain network. Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm.

[0074] The terminal 102 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal 102 and the server 101 may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present invention.

[0075] Based on the example Figure 1 In the implementation environment shown, an embodiment of the present invention provides a personnel identification method. The following is an example of the personnel identification method being applied in the terminal 102 . It can be understood that the personnel identification method can also be applied in the server 101 .

[0076] Reference Figure 2 , Figure 2 This is a flowchart of a method for identifying a person in a terminal according to an embodiment of the present invention. The subject of executing the method for identifying a person can be any of the aforementioned computer devices (including a server or a terminal). Figure 2 , the method comprises the following steps:

[0077] S100, obtaining a facial image of a target object and a feature matrix of a comparison object;

[0078] It should be noted that in some embodiments, the face image of the target object is obtained, such as Figure 3 As shown, the following steps may be included: S101, capturing a first image of a target object through a video device; the first image contains a face area of ​​the target object; S102, locating and identifying the face area of ​​the first image, and then obtaining a face image through a cropping operation.

[0079] For example, in some specific implementations, the image of the target person can be processed by a face recognition algorithm. First, the input image is analyzed to locate and identify the face area. This step ensures that the part of the image containing the face can be accurately found and framed, and then the relevant area image of this part is cropped according to the result of the algorithm calculation for subsequent detailed processing.

[0080] In some embodiments, the feature matrix of the comparison object is obtained from a preset multi-dimensional feature base library, such as Figure 4As shown, the method may also include the following steps: T100, obtaining image data of all target persons, and setting a unique identifier for each target person in a preset management library; T200, based on the image data, performing overall face area feature extraction and key part area feature extraction in turn to obtain a feature matrix for each target person; T300, associating the feature matrix corresponding to each target person with the unique identifier for regional storage, and constructing a multi-dimensional feature base library; wherein the multi-dimensional feature base library includes a first feature base library of the overall face area of ​​all target persons and a second feature base library of each key part area.

[0081] For example, in some specific implementations, establishing a multi-dimensional feature base of target personnel can be achieved as follows:

[0082] Each target person in the target person image library has a unique number, which is used to uniquely identify and manage all image data in the library. In order to build a detailed and efficient target person comprehensive feature library, multiple features are extracted from each image in the target person image library.

[0083] First, the overall features of the face are extracted through the face recognition algorithm. These features include but are not limited to facial contours, skin texture, facial proportions, etc. Next, the features of the eyes in the image are further extracted, which may include the shape, size, color, and position of the eyes. In addition, the features of the nose are also extracted, such as the shape, length, width, and position of the nose. Finally, the features of the mouth are also extracted, which include the shape, lip color, size, and position of the mouth.

[0084] After all these feature extractions are completed, a comprehensive target person feature database will be established. This feature database includes multiple base databases, namely, face feature base database, eye feature base database, nose feature base database, and mouth feature base database. The feature data in each base database will be associated with the corresponding target person number to ensure the uniqueness and accuracy of the feature data.

[0085] In addition, other characteristic attributes of the human body will be associated with the corresponding target person number. For example, the target person's height, body shape, skin color and other human characteristics may be recorded and associated. These data will also be organized into a human base feature set and associated with the target person's unique number to ensure that all characteristic data of each target person can be accurately retrieved and matched.

[0086] In this way, the comprehensive feature library of target persons realizes the comprehensive and detailed feature recording and management of each target person, so that matching and identification can be carried out quickly and accurately when performing tasks such as identity authentication and personnel tracking.

[0087] S200, extracting the face region of the face image to obtain the key face parts of the target object;

[0088] It should be noted that in some embodiments, Figure 5 As shown, performing facial region extraction on a facial image to obtain the key facial parts of a target object may include the following steps: S201, performing facial feature point detection on a facial image to obtain the key facial feature points of the target object; S202, performing regional cropping on the facial image based on the key facial feature points to obtain the key facial parts of the target object; wherein the key facial feature points represent the regional positions of the key facial parts; the key facial parts include eyes, nose and mouth.

[0089] For example, in some specific implementations, after completing the preliminary face region recognition, the cropped face image will be further analyzed in more detail using a face feature point detection algorithm. This algorithm will identify and locate key feature points on the face, including the positions of the eyes, nose, and mouth. The recognition of these feature points is based on training and learning of a large amount of face data (for example, using a residual network ResNet-10 model) to ensure that these specific areas can be accurately found.

[0090] After identifying these feature points, the image is cropped again based on the detection results to extract specific area images of the eyes, nose and mouth. These cropped specific area images can be used for a variety of purposes, such as identity authentication, expression recognition or further biometric analysis. This process ensures that the most valuable parts can be extracted from the original image for in-depth analysis and processing, thereby improving overall accuracy and reliability.

[0091] S300, extracting features from the face image and key face parts to obtain a feature point matrix;

[0092] The feature point matrix includes a first feature point matrix of the face area of ​​the target object and a second feature point matrix of the specific area of ​​each key part of the face; the feature matrix includes a first feature matrix of the face area of ​​the comparison object and a second feature matrix of the specific area;

[0093] It should be noted that in some embodiments, Figure 6 As shown, feature extraction is performed on a face image and key face parts to obtain a feature point matrix, which may include the following steps: S301, extracting a preset number of feature points from the face image and the target area of ​​each key face part in turn through a preset feature extraction algorithm; S302, based on all the feature points of each target area, correspondingly arranging to obtain a first feature point matrix of the face area and a second feature point matrix of a specific area of ​​each key face part.

[0094] For example, in some specific implementations, extracting relevant part features may be achieved as follows:

[0095] Through AI feature extraction algorithms (such as deep learning feature extraction methods based on convolutional neural networks), representative and recognizable features can be extracted from the image to generate feature vectors. For example, a matrix of 512 feature points in the relevant area can be generated:

[0096] The characteristic is β = [Q1, Q2, Q3...Q 512 ], Q h represents the hth feature point, h = [1, 2, 3…512];

[0097] The feature point matrix may include but is not limited to: face area: β1; eye area: β2; nose area: β3; mouth area: β4.

[0098] S400, according to the feature point matrix and the feature matrix, based on the distance similarity, sequentially process to obtain a similarity score of each target area of ​​the target object and the comparison object;

[0099] Among them, the target area includes the face area and various specific areas;

[0100] It should be noted that in some embodiments, Figure 7 As shown, according to the feature point matrix and the feature matrix, the similarity score of each target area of ​​the target object and the comparison object is obtained by sequentially processing based on the distance similarity, which can include the following steps: S401, according to the feature point matrix and the feature matrix, sequentially calculating the feature matrix feature difference corresponding to each target area; S402, normalizing all feature matrix feature differences to obtain the similarity score of each target area of ​​the target object and the comparison object.

[0101] For example, in some specific implementations, the face area: β1; the eye area: β2; the nose area: β3; the mouth area: β4; the feature matrix and the feature matrix of the object to be compared extracted from the database can be used to perform distance calculation, which may specifically include:

[0102] S1: Database extracts the feature matrix P of the object to be compared;

[0103] S2: Calculate the feature matrix feature difference d of different regions respectively k The calculation formula is:

[0104] Face area distance:

[0105] Eye area distance:

[0106] Nose area distance:

[0107] Mouth area distance:

[0108] S3: Normalize the minimum distance matrix of the human body and convert each distance into a score between 0 and 1:

[0109] S i =(Cos(π×(d i ÷α))+1) / 2;

[0110] α is the maximum value of the normalized score, which prevents the calculated result from being greater than 1.

[0111] The final calculated results may include: face score: S1; eye score: S2; nose score: S3; mouth score: S4.

[0112] S500: Integrate the similarity scores of all target regions to obtain target recognition scores of the target object and the comparison object.

[0113] It should be noted that, in some embodiments, the similarity scores of all target areas are integrated to obtain the target recognition score of the target object and the comparison object, which can include the following steps: based on the preset weight value of each target area, the similarity scores of all target areas are weighted summed to obtain the target recognition score of the target object and the comparison object.

[0114] For example, in some specific implementations, the final score may be calculated according to the weight, which may be implemented as follows:

[0115] After taking the aforementioned local features and calculating the individual weighted scores, these features are integrated and compared with the overall features. In order to ensure the accuracy and reliability of the matching results, the hybrid face comparison method will process each feature with different weights. Each feature has different importance in the matching process, so they are assigned corresponding weights. In this way, the system can effectively balance the influence of each feature in the recognition process, thereby ensuring that the final recognition result is more accurate and reliable.

[0116] The face segmentation algorithm is used to identify the positions of the eyes, nose, and mouth, and then their related features are extracted to calculate the distance. After normalization, the comparison results are weighted to obtain the final score:

[0117] S e =0.8×S1+0.10×S2+0.06×S3+0.04×S4;

[0118] Among them, the weights can be adaptively adjusted according to actual needs.

[0119] In order to explain the principle of the technical solution of the present invention in detail, the overall process of the present invention is described below in combination with some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and cannot be regarded as a limitation of the present invention.

[0120] First of all, it should be explained that the present invention aims to solve the problem of the singleness of existing detection technology. A personnel recognition system with mixed features is constructed through a personnel recognition method based on multi-features. The distance algorithm is combined to calculate the features of the target person and the object to be compared at different positions, and the minimum distance matrix of the face is normalized. The feature distances of the eyes, nose, and mouth are calculated, and the unique personnel recognition result that meets the requirements is output.

[0121] Among them, the hybrid face comparison method is a technology that uses high-precision matching and recognition from multiple dimensions, which has significant advantages over the traditional single-angle target person matching method. This method not only relies on the extraction and comparison of the overall features of the face, but also further refines the recognition process by segmenting the face to extract the features of the eyes, nose and mouth respectively to achieve a more accurate match.

[0122] Specifically, the hybrid face comparison method first uses advanced face recognition algorithms to process the entire face image and extract overall facial features. These features include facial contours, textures, proportions, etc., which are a comprehensive description of the entire face image. However, relying solely on these overall features is often not enough to perform high-precision person matching in complex environments.

[0123] To improve the accuracy of recognition, the hybrid method further applies a face segmentation algorithm to subdivide the entire face image into multiple regions and identify the locations of the eyes, nose, and mouth. Through this segmentation, the system can extract the features of each local area more finely. For the eyes, the system extracts features such as shape, size, color, and relative position; for the nose, the system captures information such as shape, length, width, and position; for the mouth, the system identifies features such as shape, lip color, size, and position.

[0124] After extracting these local features, the system will compare these features with the overall features. In order to ensure the accuracy and reliability of the comparison results, the hybrid face comparison method will perform weighted calculations on these features. Specifically, the matching results of different features will be given different weights according to their importance in the overall recognition. This weighted calculation method can better balance the influence of each feature in the matching process, making the final recognition result more accurate and reliable. The specific process is as follows: Figure 8 As shown, in some practical application scenarios, the technical solution of the present invention can be implemented through the following process:

[0125] 1) Establish a multi-dimensional feature base for target personnel: Each target person in the target personnel image library has a unique number, which is used to uniquely identify and manage all image data in the library. In order to build a detailed and efficient comprehensive feature base for target personnel, the system will perform multiple feature extraction on each image in the target personnel image library.

[0126] First, the system uses a face recognition algorithm to extract the overall features of the face. These features include, but are not limited to, facial contours, skin texture, and facial proportions. Next, the system further extracts the features of the eyes in the image, which may include the shape, size, color, and position of the eyes. In addition, the features of the nose are also extracted, such as the shape, length, width, and position of the nose. Finally, the system also extracts the features of the mouth, which include the shape, lip color, size, and position of the mouth.

[0127] After all these feature extractions are completed, the system will establish a comprehensive target person feature database. This feature database includes multiple base databases, namely, face feature base database, eye feature base database, nose feature base database, and mouth feature base database. The feature data in each base database will be associated with the corresponding target person number to ensure the uniqueness and accuracy of the feature data.

[0128] In addition, the system will also associate other characteristic attributes of the human body with the corresponding target person number. For example, the system may record and associate the target person's height, body shape, skin color and other human characteristics. These data will also be organized into a human base feature set and associated with the target person's unique number to ensure that all characteristic data of each target person can be accurately retrieved and matched.

[0129] In this way, the comprehensive feature library of target persons realizes the comprehensive and detailed feature recording and management of each target person, so that the system can quickly and accurately match and identify when performing tasks such as identity authentication and personnel tracking.

[0130] 2) Identify relevant parts and areas: The image of the target person is processed through the face recognition algorithm. First, the system will analyze the input image, locate and identify the face area. This step ensures that the part of the image containing the face can be accurately found and framed, and then the relevant area image of this part is cropped according to the results of the algorithm calculation for subsequent detailed processing.

[0131] After completing the initial facial region recognition, the system will further analyze the cropped facial image in more detail using a facial feature point detection algorithm. This algorithm will identify and locate key feature points on the face, including the location of the eyes, nose, and mouth. The recognition of these feature points is based on training and learning from a large amount of facial data to ensure that these specific areas can be accurately found.

[0132] After identifying these feature points, the system will crop the image again based on the detection results to extract the specific area pictures where the eyes, nose and mouth are located. These cropped specific area pictures can be used for a variety of purposes, such as identity authentication, expression recognition or further biometric analysis. This process ensures that the most valuable parts can be extracted from the original image for in-depth analysis and processing, thereby improving the overall accuracy and reliability of the system.

[0133] 3) Extract relevant part features: Through the AI ​​feature extraction algorithm, the system can extract representative and recognizable features from the image, generate feature vectors, and finally generate a matrix of 512 feature points in the relevant area:

[0134] The characteristic is β = [Q1, Q2, Q3...Q 512 ], Q h represents the hth feature point, h = [1, 2, 3…512];

[0135] The feature point matrix may include but is not limited to: face area: β1; eye area: β2; nose area: β3; mouth area: β4.

[0136] 4) Determine the feature distance and normalized score of the relevant parts: Use the face area: β1; the eye area: β2; the nose area: β3; the mouth area: β4; the feature matrix and the feature matrix of the object to be compared extracted from the database to perform distance calculation, specifically including:

[0137] S510: extracting a feature matrix P of an object to be compared from the database;

[0138] S520: Calculate the feature matrix feature difference d of different regions respectively k The calculation formula is:

[0139] Face area distance:

[0140] Eye area distance:

[0141] Nose area distance:

[0142] Mouth area distance:

[0143] S530: Perform normalization operation on the minimum distance matrix of the human body, and convert each distance into a score between 0 and 1:

[0144] S i =(Cos(π×(d i ÷α))+1) / 2;

[0145] α is the maximum value of the normalized score to prevent the calculated result from being greater than 1;

[0146] The final calculation results are: face score: S1; eye score: S2; nose score: S3; mouth score: S4.

[0147] 5) Calculate the final score based on the weights: After taking these local features and calculating the individual weighted scores, the system will integrate and compare these features with the overall features. In order to ensure the accuracy and reliability of the matching results, the hybrid face comparison method will process each feature with different weights. Each feature has different importance in the matching process, so they will be assigned corresponding weights. In this way, the system can effectively balance the influence of each feature in the recognition process, thereby ensuring that the final recognition result is more accurate and reliable.

[0148] S610: Use the face segmentation algorithm to identify the positions of the eyes, nose, and mouth respectively, extract their related features again, perform distance calculation, and perform weighted calculation on the comparison results after normalization to obtain the final score:

[0149] S e =0.8×S1+0.10×S2+0.06×S3+0.04×S4;

[0150] The above weights are based on the optimal results obtained after analyzing the existing actual scenario data. The weights can be adaptively adjusted under different application requirements.

[0151] In summary, the present invention obtains the overall facial features, eye features, nose features, and mouth features in multiple dimensions to perform high-precision matching and identification of personnel. Compared with the original target personnel matching information from a single angle, the hybrid face comparison method uses a face segmentation algorithm to identify the positions of the eyes, nose, and mouth respectively, and then extracts their related features for comparison. At the same time, the results are weighted and calculated, which further improves the accuracy of recognition. Through accurate face recognition technology, reduce the cost of manual review and improve customer experience; create new business models, such as personalized recommendations and smart advertising, which are all precision marketing based on face recognition technology; data security and privacy protection are also key. The hybrid face patent is also unique in protecting user privacy.

[0152] Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects:

[0153] 1. Establish a database of target personnel’s characteristic information from multiple angles and dimensions.

[0154] 2. Improve the accuracy of personnel identification based on the fusion of multiple features.

[0155] 3. The weighted scores of the recognition results of the three algorithms are calculated through a weighted algorithm to obtain a more accurate final comparison score.

[0156] 4. Compared with the prior art, the technical advantages of the present invention are compared as shown in Table 1 below:

[0157] Table 1

[0158]

[0159] On the other hand, Fig. 9 As shown, an embodiment of the present invention provides a person identification device 900, including:

[0160] The first module 901 is used to obtain a face image of a target object and a feature matrix of a comparison object;

[0161] The second module 902 is used to extract the face area of ​​the face image to obtain the key face parts of the target object;

[0162] The third module 903 is used to extract features from the face image and key parts of the face to obtain a feature point matrix;

[0163] The feature point matrix includes a first feature point matrix of the face area of ​​the target object and a second feature point matrix of the specific area of ​​each key part of the face; the feature matrix includes a first feature matrix of the face area of ​​the comparison object and a second feature matrix of the specific area;

[0164] The fourth module 904 is used to process the feature point matrix and the feature matrix in sequence based on the distance similarity to obtain a similarity score of each target area of ​​the target object and the comparison object; the target area includes a face area and each specific area;

[0165] The fifth module 905 is used to integrate the similarity scores of all target regions to obtain the target recognition scores of the target object and the comparison object.

[0166] Optionally, the feature matrix of the comparison object is obtained from a preset multi-dimensional feature base library, and the device may further include:

[0167] The sixth module is used to obtain the image data of all target persons and set a unique identification for each target person in the preset management library;

[0168] The seventh module is used to extract the overall face area features and key part area features in turn according to the image data to obtain the feature matrix of each target person;

[0169] The eighth module is used to associate the feature matrix corresponding to each target person with the unique identifier and store them in different regions to construct a multi-dimensional feature base library;

[0170] The multi-dimensional feature base includes a first feature base of the overall face area of ​​all target persons and a second feature base of each key part area.

[0171] The contents of the method embodiments of the present invention are all applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0172] On the other hand, an embodiment of the present invention further provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above sensitive information method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0173] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0174] like Fig.10 As shown, Fig.10 A specific example of the hardware structure of an electronic device 1000 according to an embodiment is shown. The electronic device 1000 includes:

[0175] The processor 1001 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention.

[0176] The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1002, and the processor 1001 calls and executes the network node population optimization method of the embodiment of the present invention;

[0177] Input / output interface 1003, used to implement information input and output;

[0178] The communication interface 1004 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0179] A bus 1005 , which transmits information between various components of the device (e.g., the processor 1001 , the memory 1002 , the input / output interface 1003 , and the communication interface 1004 );

[0180] The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via the bus 1005 .

[0181] The electronic device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] The contents of the method embodiments of the present invention are all applicable to the electronic device embodiments. The functions specifically implemented by the electronic device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0183] Another aspect of an embodiment of the present invention further provides a computer-readable storage medium, the storage medium stores a program, and the program is executed by a processor to implement the above method.

[0184] It should be noted that the computer-readable medium shown in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read to Only Memory, CD to ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0185] The contents of the method embodiments of the present invention are all applicable to the computer-readable storage medium embodiments. The functions specifically implemented by the computer-readable storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0186] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above method.

[0187] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0188] It should be noted that, although several modules of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0189] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD to ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation of the present invention.

[0190] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented by the present invention. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part described as a larger operation is performed independently.

[0191] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions and internal relationships of various functional modules in the device disclosed in the present invention, the actual implementation of the module will be understood within the conventional skills of engineers. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0192] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0193] The logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by an instruction execution device, apparatus or device (such as a computer-based device, a device including a processor, or other device that can fetch instructions from an instruction execution device, apparatus or device and execute the instructions), or in conjunction with such instruction execution device, apparatus or device. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution device, apparatus or device, or in conjunction with such instruction execution device, apparatus or device.

[0194] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0195] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0196] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0197] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0198] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A method for identifying a person, characterized in that: The following steps are involved: Obtaining a face image of a target object and a feature matrix of a comparison object; Extracting the facial region of the facial image to obtain the key facial parts of the target object; Extracting features from the face image and key face parts to obtain a feature point matrix; Wherein, the feature point matrix includes a first feature point matrix of the face area of ​​the target object and a second feature point matrix of the specific area of ​​each key part of the face; the feature matrix includes a first feature matrix of the face area of ​​the comparison object and a second feature matrix of the specific area; According to the feature point matrix and the feature matrix, the similarity score of each target area of ​​the target object and the comparison object is obtained by sequentially processing based on the distance similarity; the target area includes the face area and each of the specific areas; The similarity scores of all the target regions are integrated to obtain target recognition scores of the target object and the comparison object.

2. The method for identifying a person according to claim 1, characterized in that: The step of obtaining the face image of the target object comprises the following steps: Capturing a first picture of the target object through a video device; the first picture contains a face area of ​​the target object; The face area of ​​the first picture is located and identified, and then the face image is obtained through a cropping operation.

3. The method for identifying a person according to claim 1, characterized in that: The feature matrix of the comparison object is obtained from a preset multi-dimensional feature base library, and the method further includes the following steps: Acquire image data of all target persons, and set a unique identifier for each of the target persons in a preset management library; Based on the image data, extracting features of the entire face region and extracting features of key parts of the face region in sequence to obtain the feature matrix of each target person; The feature matrix corresponding to each target person is associated with the unique identifier and stored in different regions to construct the multi-dimensional feature base library; The multi-dimensional feature base includes a first feature base of the overall face area of ​​all target persons and a second feature base of each key part area.

4. The method for identifying a person according to claim 1, characterized in that: The step of extracting the face region from the face image to obtain the key face parts of the target object comprises the following steps: Performing facial feature point detection on the facial image to obtain key facial feature points of the target object; Performing region cropping on the face image based on the key facial feature points to obtain the key facial parts of the target object; Among them, the key feature points of the face represent the regional positions of the key parts of the face; the key parts of the face include eyes, nose and mouth.

5. The method for identifying a person according to claim 1, characterized in that: The step of extracting features from the face image and the key parts of the face to obtain a feature point matrix comprises the following steps: Extracting a preset number of feature points from the face image and the target area of ​​each key part of the face in sequence through a preset feature extraction algorithm; Based on all the feature points of each target area, the first feature point matrix of the face area and the second feature point matrix of the specific area of ​​each key part of the face are correspondingly sorted out.

6. The method for identifying a person according to claim 1, characterized in that: The method of obtaining a similarity score between each target area of ​​the target object and the comparison object by sequentially processing the feature point matrix and the feature matrix based on distance similarity comprises the following steps: According to the feature point matrix and the feature matrix, sequentially calculating the feature matrix feature difference value corresponding to each of the target areas; A normalization operation is performed on all the feature matrix feature differences to obtain the similarity score of each target region of the target object and the comparison object.

7. The method for identifying a person according to claim 1, characterized in that: The integrating operation of the similarity scores of all the target regions to obtain the target recognition scores of the target object and the comparison object comprises the following steps: Based on the preset weight value of each target area, the similarity scores of all the target areas are weighted and summed to obtain the target recognition score of the target object and the comparison object.

8. A person identification device, characterized in that: include: The first module is used to obtain the face image of the target object and the feature matrix of the comparison object; The second module is used to extract the face area of ​​the face image to obtain the key face parts of the target object; The third module is used to extract features from the face image and key parts of the face to obtain a feature point matrix; Wherein, the feature point matrix includes a first feature point matrix of the face area of ​​the target object and a second feature point matrix of the specific area of ​​each key part of the face; the feature matrix includes a first feature matrix of the face area of ​​the comparison object and a second feature matrix of the specific area; A fourth module is used to sequentially process the feature point matrix and the feature matrix based on distance similarity to obtain similarity scores of each target area of ​​the target object and the comparison object; the target area includes the face area and each of the specific areas; The fifth module is used to integrate the similarity scores of all the target areas to obtain the target recognition scores of the target object and the comparison object.

9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 7.

10. A computer storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the method according to any one of claims 1 to 7 when executed by the processor.

Images