Finger recognition method and device, storage medium and electronic device

Abstract

This disclosure relates to a finger recognition method, apparatus, storage medium, and electronic device. The method includes: acquiring an image to be recognized, the image including a finger to be recognized; extracting features from the finger to be recognized in the image to generate a feature vector; matching the feature vector with a preset feature vector database to determine the finger type; the preset feature vector database includes a one-to-one correspondence between multiple feature vectors and multiple finger types. This method extracts features from the finger to be recognized in the image, generates a feature vector corresponding to the finger, and matches the feature vector with the preset feature vector database to determine the finger type. By recognizing the finger based on its features, the method can accurately identify the finger type even when the finger image is incomplete, improving the reliability and accuracy of finger type recognition.

Description

Technical Field

[0001] This disclosure relates to image recognition, and more specifically, to a finger recognition method, apparatus, storage medium, and electronic device. Background Technology

[0002] In related technologies, the type of fingers a user is displaying can be determined by recognizing the user's palm in an image. However, in many visual scenarios, due to the limitations of the camera's field of view, there are situations where the user's palm is obscured, making it impossible to identify the finger type. For example, when the user's palm is obscured, only a portion of the finger image can be identified, making it impossible to determine the finger type based on the arrangement of the fingers in the palm. Summary of the Invention

[0003] The purpose of this disclosure is to provide a finger recognition method, apparatus, storage medium, and electronic device to solve the technical problem in the prior art that it is impossible to accurately identify the type of finger from an image.

[0004] According to a first aspect of this disclosure, a finger recognition method is provided, the method comprising:

[0005] Acquire an image to be identified, wherein the image to be identified includes a finger to be identified;

[0006] Feature extraction is performed on the finger to be identified in the image to be identified, so as to generate a feature vector of the finger to be identified;

[0007] The feature vector is matched with a preset feature vector database to determine the finger type of the finger to be identified. The preset feature vector database includes a one-to-one correspondence between multiple feature vectors and multiple finger types.

[0008] Optionally, the step of extracting features from the finger to be identified in the image to generate a feature vector of the finger to be identified includes:

[0009] The finger to be identified is identified in order to determine the deflection angle of the finger in a preset direction;

[0010] Based on the deflection angle, feature extraction is performed on the finger to be identified to generate the feature vector of the finger to be identified at the deflection angle.

[0011] Optionally, matching the feature vector with a preset feature vector database to determine the finger type of the finger to be identified includes:

[0012] Determine the target feature vector data corresponding to the deflection angle from the preset feature vector database;

[0013] The feature vector is matched with the target feature vector data to determine the finger type of the finger to be identified.

[0014] Optionally, the step of identifying the finger to be identified to determine the deflection angle of the finger to be identified in a preset direction includes:

[0015] The finger to be identified in the image to be identified is identified in order to determine the fingernail outline of the finger to be identified;

[0016] Based on the fingernail outline diagram, the rotation angle of the finger to be identified in a preset direction is determined.

[0017] Optionally, the method further includes:

[0018] Obtain the user identifier corresponding to the image to be identified;

[0019] The preset feature vector database is determined based on the user identifier.

[0020] Optionally, the method further includes:

[0021] If it is determined that a complete palm image corresponding to the finger to be identified exists in the image to be identified, the finger to be identified is identified based on the complete palm image to determine the target finger type of the finger to be identified.

[0022] Optionally, the method further includes:

[0023] Multiple image features are generated by extracting features from multiple finger images at multiple preset adjacent angles using a preset feature extraction network.

[0024] The multiple image features are fused according to a preset feature fusion network to generate multiple feature vectors;

[0025] The preset feature vector database is generated based on the multiple feature vectors.

[0026] According to a second aspect of this disclosure, a finger recognition device is provided, the device comprising:

[0027] The acquisition module is used to acquire an image to be identified, wherein the image to be identified includes a finger to be identified;

[0028] The generation module is used to extract features from the finger to be identified in the image to be identified, so as to generate a feature vector of the finger to be identified;

[0029] The determination module is used to match the feature vector with a preset feature vector database to determine the finger type of the finger to be identified. The preset feature vector database includes a one-to-one correspondence between multiple feature vectors and multiple finger types.

[0030] According to a third aspect of this disclosure, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described in any of the first aspects of this disclosure.

[0031] According to a fourth aspect of this disclosure, an electronic device is provided, comprising:

[0032] A memory on which computer programs are stored;

[0033] A processor for executing the computer program in the memory to implement the steps of the method according to any one of the first aspects of this disclosure.

[0034] The above technical solution acquires an image to be identified, which includes a finger to be identified. Features of the finger are extracted from the image to generate a feature vector. This feature vector is then matched with a pre-set feature vector database to determine the finger type. The database contains a one-to-one correspondence between multiple feature vectors and multiple finger types. This method, which identifies the finger based on its features, can accurately identify the finger type even when the finger image is incomplete, thus improving the reliability and accuracy of finger type recognition.

[0035] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0036] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings:

[0037] Figure 1 This is a flowchart illustrating a finger recognition method according to an exemplary embodiment;

[0038] Figure 2 This is an example diagram illustrating an image to be identified according to an exemplary embodiment;

[0039] Figure 3 This is a flowchart illustrating a finger recognition method according to an exemplary embodiment;

[0040] Figure 4 This is a schematic diagram illustrating a finger detection method according to an exemplary embodiment;

[0041] Figure 5 This is a flowchart illustrating a finger detection method according to an exemplary embodiment;

[0042] Figure 6 This is an example diagram illustrating a finger type recognition method according to an exemplary embodiment;

[0043] Figure 7 This is an example diagram illustrating a method for generating a preset feature vector database according to an exemplary embodiment;

[0044] Figure 8 This is a flowchart illustrating a finger recognition method according to an exemplary embodiment;

[0045] Figure 9 This is a block diagram illustrating a finger recognition device according to an exemplary embodiment;

[0046] Figure 10 This is a block diagram illustrating an electronic device according to an exemplary embodiment;

[0047] Figure 11 This is a block diagram illustrating a network device according to an exemplary embodiment. Detailed Implementation

[0048] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0049] It should be noted that all actions involving the acquisition of signals, information, or data in this disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with authorization from the owner of the relevant device.

[0050] Figure 1 This is a flowchart illustrating a finger recognition method according to an exemplary embodiment, such as... Figure 1 As shown, the method includes the following steps.

[0051] Step S101: Obtain the image to be recognized, which includes the finger to be recognized.

[0052] It is worth mentioning that this embodiment is applied to a terminal, which can be a user terminal or a network terminal. The terminal is used to identify the finger to be identified in the image to be identified, thereby determining the finger type, which includes the thumb, index finger, or middle finger, etc. An image to be identified is input into the terminal, enabling the terminal to perform image recognition. Based on the image features of the finger, the image corresponding to the finger to be identified can be extracted from the image to be identified.

[0053] Optionally, in one embodiment, after step S101 above, the identification method includes:

[0054] If a complete palm image corresponding to the finger to be identified exists in the image to be identified, the finger to be identified is identified based on the complete palm image to determine the target finger type.

[0055] For example, in this embodiment, the identification of a finger in an image to be identified is performed. When the user's finger shown in the image is incomplete, for example, only three fingers or only one finger are visible, the inherent arrangement pattern of fingers cannot be used to distinguish the finger type. However, by identifying a complete palm image corresponding to the finger in the image to be identified, the target finger type can be determined based on the arrangement pattern of the finger in the complete palm image. For example, Figure 2 This is an example diagram illustrating an image to be recognized according to an exemplary embodiment, such as... Figure 2 As shown, when identifying finger a through the above steps, a complete palm image corresponding to the finger to be identified can be determined from a. Then, based on the arrangement pattern of the finger to be identified in a, the finger to be identified in a can be directly identified, and the finger to be identified in a is determined to be: thumb, index finger, middle finger, ring finger, and little finger. However, when identifying the finger to be identified in b, since a complete palm image does not exist in the image to be identified, it is necessary to extract features from the finger to be identified in b and determine the finger type of the finger to be identified through feature vectors.

[0056] Step S102: Extract features from the finger to be identified in the image to generate a feature vector of the finger to be identified.

[0057] For example, in this embodiment, based on the features of the finger to be identified in the image, such as the texture features of the finger, the length features between each phalanx, and the line features on the phalanx, feature extraction is performed on the image corresponding to the finger to be identified to generate a feature vector of the finger to be identified. For example, a trained feature extraction network can be used to extract features of the finger to be identified and generate a feature vector based on the proposed features. It is worth mentioning that, due to the special characteristics of users' fingers, the features of the fingers to be identified are different for different users. Therefore, when the same feature extraction network is used to extract features from the same finger of the same user, the obtained feature vector is the same, and the features of different types of fingers of the same user are different. Therefore, the finger type can be determined based on the feature vector of the finger.

[0058] Step S103: Match the feature vector with a preset feature vector database to determine the finger type of the finger to be identified. The preset feature vector database includes a one-to-one correspondence between multiple feature vectors and multiple finger types.

[0059] It is worth mentioning that in this embodiment, feature data is recorded for each of the user's fingers to generate a preset feature vector database. This database includes multiple feature vectors corresponding to multiple finger types, with each finger type corresponding to a specific feature vector. The feature vectors determined in the above steps are compared with the multiple feature vectors pre-recorded in the preset feature vector database. The pre-recorded feature vectors that match the feature vectors are selected from the database. Based on the correspondence between the feature vectors and finger types in the database, the finger type corresponding to the finger to be identified is determined.

[0060] Alternatively, in another embodiment, before step S103 above, the identification method further includes:

[0061] Obtain the user identifier corresponding to the image to be identified.

[0062] The preset feature vector database is determined based on the user identifier.

[0063] For example, in this embodiment, the recognition accuracy of the feature vectors obtained through the above steps is related to the amount of data on the feature vectors corresponding to the user's fingers and finger types pre-recorded in the preset feature vector database. However, when there are many feature vectors in the preset feature vector database, the recognition efficiency of the finger to be recognized will decrease. Therefore, in order to improve the accuracy of the recognition of the finger type corresponding to the finger to be recognized, as well as the recognition efficiency of the finger to be recognized, in this embodiment, the image to be recognized is used to recognize the finger type of a specific user. The preset feature vector database pre-records the finger features and corresponding finger types of each finger of the specific user. After determining the feature vectors corresponding to the finger to be recognized in the image to be recognized through the above steps, the user identifier of the corresponding user is determined by recognizing the image to be recognized. Based on the user identifier, the target feature vector data is determined from the preset feature vector database. The target feature vector data stores the feature vectors corresponding to each finger of the user and the corresponding finger types. The target feature vector data is compared with the feature vectors obtained in the above steps to determine the finger type corresponding to the finger to be recognized.

[0064] The above technical solution acquires an image to be identified, which includes a finger to be identified. Features of the finger are extracted from the image to generate a feature vector. This feature vector is then matched with a pre-set feature vector database to determine the finger type. The database contains a one-to-one correspondence between multiple feature vectors and multiple finger types. This method, which identifies the finger based on its features, can accurately identify the finger type even when the finger image is incomplete, thus improving the reliability and accuracy of finger type recognition.

[0065] Figure 3 This is a flowchart illustrating a finger recognition method according to an exemplary embodiment, such as... Figure 3 As shown, the method includes the following steps.

[0066] Step S201: Obtain the image to be recognized, which includes the finger to be recognized.

[0067] For example, in this embodiment, the method of obtaining the image to be identified is the same as in step S101 above, and can be referred to step S101 above, without further description.

[0068] In one embodiment, step S101 above may include steps S201 and S202.

[0069] Step S202: Identify the finger to be identified to determine the deflection angle of the finger in a preset direction.

[0070] It is worth mentioning that when extracting features from the same finger using the same feature extraction network, the texture and length features of the finger in the image to be identified differ depending on the angle of the finger. Therefore, in this embodiment, feature extraction is performed on the finger to be identified based on the deflection angle of the finger in the image. For example, Figure 4 This is a schematic diagram illustrating a finger detection method according to an exemplary embodiment, such as... Figure 4 As shown, in this embodiment, when identifying a finger in an image, the fingernail is identified. Using the horizontal plane as a reference plane, the rolling angle of the plane containing the fingernail relative to the reference plane is determined, which is then used as the deflection angle of the finger in a preset direction. For example, Figure 5 This is a flowchart illustrating a finger detection method according to an exemplary embodiment, such as... Figure 5 As shown, the method includes inputting an image to be identified into an encoding module, wherein the image to be identified includes a finger to be identified; after encoding the image to be identified by the encoding module, the encoding result is input into a decoding module; wherein the encoding module includes a feature recognition neural network; the feature recognition neural network identifies the features of the finger to be identified in the image to be identified, and the recognition result is input into the decoding module; the decoding module includes two deconvolution layers; the deconvolution layers extract features from the recognition result to generate a segmented image of the nail contour corresponding to the finger to be identified, as well as the deflection angle of the finger in the Z-axis direction.

[0071] Step S203: Extract features of the finger to be identified based on the deflection angle to generate a feature vector of the finger to be identified at the deflection angle.

[0072] For example, the feature vector extraction method in this embodiment is the same as in step S102 above, and can be referred to step S102 above, without further description.

[0073] Step S204: Match the feature vector with a preset feature vector database to determine the finger type of the finger to be identified. The preset feature vector database includes a one-to-one correspondence between multiple feature vectors and multiple finger types.

[0074] For example, the method for determining the finger type in this embodiment is the same as in step S103 above, and can be referred to step S103 above, so it will not be repeated here.

[0075] Alternatively, in another embodiment, step S204 above includes:

[0076] Determine the target feature vector data corresponding to the deflection angle from the preset feature vector database.

[0077] The feature vector is matched with the target feature vector data to determine the finger type of the finger to be identified.

[0078] For example, in this embodiment, after obtaining the feature vector of the finger to be identified at the corresponding deflection angle through the above steps, the target feature vector data corresponding to the user identifier corresponding to the image to be identified is determined from a preset feature vector database. This target feature vector data includes feature vectors of each finger of the user corresponding to the user identifier in each direction. Based on the deflection angle, the target feature vector data is filtered to obtain multiple feature vectors corresponding to each finger at that deflection angle. The feature vectors determined through the above steps are compared with these multiple feature vectors to determine the finger type corresponding to the finger to be identified. For example, Figure 6 This is an example diagram illustrating a finger type recognition method according to an exemplary embodiment, such as... Figure 6 As shown, the method includes an input stage and a discrimination stage. In the input stage, feature vectors corresponding to each finger of the user at the same angle are input and stored in a preset feature vector database. In the discrimination stage, after identifying the deflection angle of the finger to be identified through the above steps, the feature vectors corresponding to each finger are compared with the feature vectors of the finger to be identified to determine the finger type. For example, the feature vectors corresponding to each finger of user A are pre-input into the preset feature vector database. For each finger of user A, four feature vectors need to be input within four angle ranges: -40° to -20°, -20° to 0°, 0° to 20°, and 20° to 40°. After determining that the user corresponding to the image to be identified is user A, and that the deflection angle of the finger to be identified in the image is 30°, the feature vectors are compared with the feature vectors of each finger of user A within the 20° to 40° angle range to determine the finger type.

[0079] Alternatively, in another embodiment, before step S204 above, the identification method further includes:

[0080] Multiple image features are generated by extracting features from multiple finger images at multiple preset adjacent angles using a preset feature extraction network.

[0081] Multiple image features are fused using a pre-defined feature fusion network to generate multiple feature vectors.

[0082] A preset feature vector database is generated based on multiple feature vectors.

[0083] It is worth mentioning that before comparing the feature vectors corresponding to the fingers to be identified, it is necessary to input images of each finger of the user corresponding to the image to be identified, and generate feature vectors corresponding to each finger based on the input finger images. In this embodiment, a preset feature extraction network is used to extract features from multiple finger images at preset adjacent angles, generating multiple image features corresponding to each finger. Multiple image features are then fused using a preset feature fusion network to generate multiple feature vectors, and a preset feature vector database is generated based on these multiple feature vectors. For example, Figure 7 This is an example diagram illustrating a method for generating a preset feature vector database according to an exemplary embodiment, such as... Figure 7 As shown, the input angle range 0 is set to -40° to -20°, input angle range 1 is set to -20° to 20°, and input angle range 2 is set to 20° to 40°. A preset feature extraction network module extracts features from multiple finger images within the same input angle range for the same finger, generating multiple image features. For example, for input angle range 0, multiple image features of the user's index finger at -23°, -30°, and -35° can be collected. Then, a feature fusion network module fuses these multiple image features to generate a feature vector for the finger within input angle range 0. This process iterates through the user's fingers, obtaining multiple feature vectors, and a preset feature vector database is generated based on these feature vectors.

[0084] It is worth mentioning that in this embodiment, a loss function is used to train a preset feature extraction network and a preset feature fusion network to generate a feature extraction model. Based on this feature extraction model, features of the finger to be identified are extracted during the input and discrimination stages to generate a feature vector. The loss function is:

[0085] loss1=|F0-F1|1-|F1-F2|2-|F0-F2|1

[0086] loss2=KL(F0-F1)-KL(F1-F2)-KL(F0-F1)

[0087] loss=αloss1+βloss2, α<β

[0088] The loss function consists of two main parts: loss1 and loss2. Loss1 primarily calculates the L1 distance between features, while loss2 primarily calculates the KL divergence similarity between features. Based on the loss function, the main purpose of training this model is to ensure that feature vectors with similar angles are as similar as possible, ultimately making the loss α less than β. The KL divergence reflects the overall distribution differences of the features. For example, in this embodiment, a transformer network structure can be selected as the encoder (feature fusion network). This transformer network structure is an encoder-decode framework, i.e., an encoding-decoding network architecture. The encoder network is pre-trained on a dataset with more than 5000 classes to ensure better training of the transformer module. After pre-training, the encoder network is further fine-tuned using the loss function to ensure the final result, generating a feature extraction model. This feature extraction model is used to extract features from the finger to be identified in the image to generate the corresponding feature vector.

[0089] The above technical solution acquires an image to be identified, which includes a finger to be identified. Features of the finger are extracted from the image to generate a feature vector. This feature vector is then matched with a pre-set feature vector database to determine the finger type. The database contains a one-to-one correspondence between multiple feature vectors and multiple finger types. This method, which identifies the finger based on its features, can accurately identify the finger type even when the finger image is incomplete, thus improving the reliability and accuracy of finger type recognition.

[0090] Figure 8 This is a flowchart illustrating a finger recognition method according to an exemplary embodiment, such as... Figure 8 As shown, the method includes the following steps.

[0091] I. Data Entry Stage

[0092] (1) Input images of each of the user's fingers;

[0093] (2) Based on the finger type of each finger, input the finger image of the corresponding finger of the user;

[0094] (3) Extract features from the finger image to obtain the feature vector of the user's corresponding finger, and store the feature vector in the database;

[0095] II. Judgment Stage

[0096] (1) Input the image to be recognized;

[0097] (2) Detect the finger to be identified in the image to be identified, and determine whether the entire palm image can be detected in the image to be identified. If the entire palm image exists in the image to be identified, directly output the finger type of the finger to be identified.

[0098] (3) If the entire palm image cannot be detected in the image to be identified, the finger to be identified in the image to be identified is detected, the features of the finger to be identified are extracted, the feature vector of the finger to be identified is generated, and the feature vector is matched with the feature vector in the database to determine the finger type of the finger to be identified.

[0099] Figure 9 This is a block diagram illustrating a finger recognition device according to an exemplary embodiment, such as... Figure 9 As shown, the device 100 includes: an acquisition module 110, a generation module 120, and a determination module 130.

[0100] The acquisition module 110 is used to acquire an image to be recognized, which includes a finger to be recognized.

[0101] The generation module 120 is used to extract features from the finger to be identified in the image to generate a feature vector of the finger to be identified.

[0102] The determination module 130 is used to match the feature vector with a preset feature vector database to determine the finger type of the finger to be identified. The preset feature vector database includes a one-to-one correspondence between multiple feature vectors and multiple finger types.

[0103] Optionally, the generation module 120 includes:

[0104] The determination submodule is used to identify the finger to be identified in order to determine the deflection angle of the finger in a preset direction.

[0105] The generation submodule is used to extract features of the finger to be identified based on the deflection angle, so as to generate the feature vector of the finger to be identified at the deflection angle.

[0106] Optionally, module 130 is defined for:

[0107] Determine the target feature vector data corresponding to the deflection angle from the preset feature vector database.

[0108] The feature vector is matched with the target feature vector data to determine the finger type of the finger to be identified.

[0109] Optionally, a submodule is defined for:

[0110] The process involves identifying the finger in the image to be identified in order to determine the outline of the fingernail.

[0111] Based on the fingernail outline, determine the rotation angle of the finger to be identified in a preset direction.

[0112] Optionally, the device 100 includes an acquisition module, which is used for:

[0113] Obtain the user identifier corresponding to the image to be identified.

[0114] The preset feature vector database is determined based on the user identifier.

[0115] Optionally, the device 100 includes a determination module, which is used for:

[0116] If a complete palm image corresponding to the finger to be identified exists in the image to be identified, the finger to be identified is identified based on the complete palm image to determine the target finger type.

[0117] Optionally, the device 100 includes an execution module, which is used for:

[0118] Multiple image features are generated by extracting features from multiple finger images at multiple preset adjacent angles using a preset feature extraction network.

[0119] Multiple image features are fused using a pre-defined feature fusion network to generate multiple feature vectors.

[0120] A preset feature vector database is generated based on multiple feature vectors.

[0121] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0122] Figure 10 This is a block diagram illustrating an electronic device 1000 according to an exemplary embodiment. For example... Figure 10 As shown, the electronic device 1000 may include: a first processor 1001 and a first memory 1002. The electronic device 1000 may also include one or more of the following: a multimedia component 1003, a first input / output (I / O) interface 1004, and a first communication component 1005.

[0123] The first processor 1001 controls the overall operation of the electronic device 1000 to complete all or part of the steps in the aforementioned finger recognition method. The first memory 1002 stores various types of data to support the operation of the electronic device 1000. This data may include, for example, instructions for any application or method operating on the electronic device 1000, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The first memory 1002 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. Multimedia component 1003 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the first memory 1002 or transmitted via the first communication component 1005. The audio component also includes at least one speaker for outputting audio signals. First I / O interface 1004 provides an interface between the first processor 1001 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. First communication component 1005 is used for wired or wireless communication between the electronic device 1000 and other devices. Wireless communication includes, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IoT, eMTC, or other 5G technologies, or combinations thereof, and is not limited herein. Therefore, the corresponding first communication component 1005 may include: a Wi-Fi module, a Bluetooth module, an NFC module, etc.

[0124] In an exemplary embodiment, the electronic device 1000 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the finger recognition method described above.

[0125] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the finger recognition method described above. For example, the computer-readable storage medium may be the first memory 1002 including the program instructions, which may be executed by the first processor 1001 of the electronic device 1000 to complete the finger recognition method described above.

[0126] Figure 11 This is a block diagram illustrating a network device 1100 according to an exemplary embodiment. For example, network device 1100 may be provided as a server. (Refer to...) Figure 11 The network device 1100 includes a second processor 1122, which may be one or more, and a second memory 1132 for storing computer programs executable by the second processor 1122. The computer program stored in the second memory 1132 may include one or more modules, each corresponding to a set of instructions. Furthermore, the second processor 1122 may be configured to execute the computer program to perform the aforementioned finger recognition method.

[0127] Additionally, network device 1100 may also include a power supply component 1126 and a second communication component 1150. The power supply component 1126 may be configured to perform power management of network device 1100, and the second communication component 1150 may be configured to enable communication of network device 1100, such as wired or wireless communication. Furthermore, network device 1100 may also include a second input / output (I / O) interface 1158. Network device 1100 can operate on an operating system stored in a second memory 1132.

[0128] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the finger recognition method described above. For example, the non-transitory computer-readable storage medium may be the second memory 1132 including the program instructions, which may be executed by the second processor 1122 of the network device 1100 to complete the finger recognition method described above.

[0129] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the above-described finger recognition method when executed by the programmable device.

[0130] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0131] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction.

[0132] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A method of finger recognition, characterized by, The method comprises: acquiring a to-be-identified image, the to-be-identified image comprising a to-be-identified finger; identifying the to-be-identified finger in the to-be-identified image to determine a fingernail contour graph of the to-be-identified finger; determining a deflection angle of the to-be-identified finger in a preset direction according to the fingernail contour graph; extracting features of the to-be-identified finger according to the deflection angle to generate a feature vector of the to-be-identified finger at the deflection angle; determining target feature vector data corresponding to the deflection angle from a preset feature vector database; matching the feature vector with the target feature vector data to determine a finger type of the to-be-identified finger, the preset feature vector database comprising a one-to-one correspondence relationship between a plurality of feature vectors and a plurality of finger types; The method further comprises: extracting features of a plurality of finger images at a plurality of preset adjacent angles through a preset feature extraction network to generate a plurality of image features; fusing the plurality of image features according to a preset feature fusion network to generate a plurality of feature vectors; generating the preset feature vector database according to the plurality of feature vectors.

2. The identification method according to claim 1, characterized in that, The method further comprises: acquiring a user identifier corresponding to the to-be-identified image; determining the preset feature vector database according to the user identifier.

3. The identification method according to claim 1, characterized in that, The method further comprises: in a case where it is determined that a complete palm image corresponding to the to-be-identified finger exists in the to-be-identified image, identifying the to-be-identified finger according to the complete palm image to determine a target finger type of the to-be-identified finger.

4. A finger identification device, characterized by The device comprises: an acquisition module configured to acquire a to-be-identified image, the to-be-identified image comprising a to-be-identified finger; a generation module configured to identify the to-be-identified finger in the to-be-identified image to determine a fingernail contour graph of the to-be-identified finger, determine a deflection angle of the to-be-identified finger in a preset direction according to the fingernail contour graph, and extract features of the to-be-identified finger according to the deflection angle to generate a feature vector of the to-be-identified finger at the deflection angle; a determination module configured to determine target feature vector data corresponding to the deflection angle from a preset feature vector database, and match the feature vector with the target feature vector data to determine a finger type of the to-be-identified finger, the preset feature vector database comprising a one-to-one correspondence relationship between a plurality of feature vectors and a plurality of finger types; an execution module configured to: extract features of a plurality of finger images at a plurality of preset adjacent angles through a preset feature extraction network to generate a plurality of image features; fuse the plurality of image features according to a preset feature fusion network to generate a plurality of feature vectors; generate the preset feature vector database according to the plurality of feature vectors.

5. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The program, when executed by a processor, implements the steps of the method of any one of claims 1-3.

6. An electronic device, comprising: comprises: a memory having a computer program stored thereon; a processor configured to execute the computer program in the memory to implement the steps of the method of any one of claims 1-3.

Images