Training methods for palmprint recognition models, palmprint recognition methods and related equipment

By preprocessing palmprint images from different devices and constructing positive and negative sample pairs, a palmprint recognition model is trained, solving the problem of cross-device recognition accuracy and achieving efficient recognition across multiple devices.

CN118982848BActive Publication Date: 2026-11-17PING AN TECH (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411185606.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2026-11-17
Estimated Expiration
2044-08-27

AI Technical Summary

Technical Problem

Existing palmprint recognition technology has failed to effectively optimize for differences between different devices, resulting in decreased accuracy when recognizing palmprints across devices.

Method used

By acquiring palmprint images from different devices, preprocessing them based on device type, constructing positive and negative sample pairs, and using a contrastive learning model to train the palmprint recognition model, cross-device adaptability is improved.

Benefits of technology

It improves the accuracy and robustness of cross-device palmprint recognition, adapts to the imaging conditions of different devices, and ensures the reliability of recognition in various environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982848B_ABST
    Figure CN118982848B_ABST
Patent Text Reader

Abstract

This invention discloses a training method for a palmprint recognition model, a palmprint recognition method, and related equipment. The training method includes: acquiring palmprint images collected from different devices, and preprocessing the palmprint images according to the type of the acquisition device; inputting the preprocessed palmprint images into a feature extraction network to extract image features from the palmprint images; constructing positive and negative sample pairs for the image features based on the user and acquisition device of the palmprint images to obtain a training sample set; inputting the training sample set into a pre-constructed contrastive learning model for training until a preset convergence condition is met, thus completing the model training and obtaining a palmprint recognition model. By using different feature processing methods for palmprint images collected from different devices and constructing positive and negative sample pairs for contrastive learning, the trained palmprint recognition model has stronger cross-device adaptability and improves the accuracy of palmprint recognition across devices.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of financial technology, and in particular to a training method for palmprint recognition models, a palmprint recognition method, and related equipment. Background Technology

[0002] Palmprint recognition already has some similar products and applications in the financial sector. For example, financial institutions have begun using palmprint recognition technology as a means of identity verification and transaction authentication. In addition, some mobile payment platforms have also begun to introduce palmprint payment as a convenient payment method.

[0003] In palmprint payment and financial transaction scenarios, there may be situations where users register their palmprints on a mobile device, but the actual verification scenario is not on a mobile device. Device differences may lead to inconsistent palmprint image acquisition conditions. For example, factors such as lighting intensity, angle, and image quality between different devices may result in inconsistent palmprint image quality. Existing palmprint recognition technology has not been optimized for the problems that may arise from differences between different devices. This means that when performing palmprint recognition across devices, device differences may affect palmprint recognition, thus impacting the reliability of palmprint recognition. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this invention is to provide a training method, a palm print recognition method and related equipment for palm print recognition models that can be applied to financial technology or other related fields, with the main purpose of improving the accuracy of palm print recognition across devices.

[0005] The technical solution of the present invention is as follows:

[0006] The first aspect of this invention provides a method for training a palmprint recognition model, comprising:

[0007] Acquire palmprint images from different devices and perform corresponding preprocessing on the palmprint images based on the type of acquisition device;

[0008] The preprocessed palmprint image is input into a feature extraction network to extract the image features of the palmprint image;

[0009] Based on the user and acquisition device of the palmprint image, positive and negative sample pairs are constructed from the image features to obtain a training sample set;

[0010] The training sample set is input into a pre-built contrastive learning model for training until a preset convergence condition is met, at which point the model training is complete, and a palmprint recognition model is obtained.

[0011] In one embodiment, the preprocessing of the palmprint image based on the type of acquisition device includes:

[0012] Confirm whether the device used to acquire the palmprint image is a mobile terminal;

[0013] When the acquisition device is a mobile terminal, the palmprint image is preprocessed according to the first preprocessing strategy;

[0014] When the acquisition device is not a mobile terminal, the palmprint image is preprocessed according to the second preprocessing strategy.

[0015] In one embodiment, preprocessing the palmprint image according to the first preprocessing strategy includes:

[0016] The size of the palm print image is adjusted to fit the target size of the mobile terminal;

[0017] The brightness and contrast of the palmprint image are adaptively adjusted.

[0018] The adaptively adjusted palmprint image is then subjected to noise reduction processing.

[0019] In one embodiment, preprocessing the palmprint image according to the second preprocessing strategy includes:

[0020] The size of the palmprint image is adjusted to match the target size of the mobile terminal;

[0021] The palmprint image is subjected to histogram equalization and its color space is converted to a specified color space.

[0022] In one embodiment, the step of constructing a training sample set by pairing positive and negative samples based on the user and acquisition device of the palmprint image includes:

[0023] Positive sample pairs are constructed by using image features collected from the same user on different devices as positive samples.

[0024] Select negative samples that are different from the positive samples from image features collected by other users on any device, and construct negative sample pairs with the positive samples;

[0025] The positive and negative sample pairs are combined to form a positive-negative sample pair, thus obtaining the training sample set.

[0026] A second aspect of the present invention provides a palmprint recognition method, comprising:

[0027] Acquire a palmprint image to be identified, and extract the image features of the palmprint image to be identified;

[0028] The image features of the palmprint image to be identified are input into a pre-trained palmprint recognition model, and the palmprint is compared and recognized with the palmprint database of registered users. The palmprint recognition result of the palmprint image to be identified is then output.

[0029] The pre-trained palmprint recognition model was obtained using the training method described above.

[0030] A third aspect of the present invention provides a training apparatus for a palmprint recognition model, comprising:

[0031] The data collection and processing module acquires palmprint images collected from different devices and performs corresponding preprocessing on the palmprint images based on the type of acquisition device.

[0032] The feature extraction module is used to input the preprocessed palmprint image into the feature extraction network to extract the image features of the palmprint image;

[0033] The sample construction module is used to construct positive and negative sample pairs for the image features based on the user and acquisition device to which the palmprint image belongs, thereby obtaining a training sample set;

[0034] The model training module is used to input the training sample set into a pre-built contrastive learning model for training until a preset convergence condition is met, thus completing the model training and obtaining a palmprint recognition model.

[0035] A fourth aspect of the present invention provides a palmprint recognition device, comprising:

[0036] The image acquisition module is used to acquire a palmprint image to be identified and extract the image features of the palmprint image to be identified;

[0037] The recognition module is used to input the image features of the palmprint image to be recognized into a pre-trained palmprint recognition model, compare and recognize the palmprint with the palmprint database of registered users, and output the palmprint recognition result of the palmprint image to be recognized.

[0038] The pre-trained palmprint recognition model was obtained using the training method described above.

[0039] A fifth aspect of the present invention provides an electronic device comprising at least one processor; and,

[0040] A memory communicatively connected to the at least one processor; wherein,

[0041] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the training method of the palmprint recognition model or to perform the palmprint recognition method described above.

[0042] A sixth aspect of the present invention provides a non-volatile computer-readable storage medium storing computer-executable instructions, which, when executed by one or more processors, cause the one or more processors to perform the above-described palmprint recognition model training method or to perform the above-described palmprint recognition method.

[0043] Beneficial Effects: This invention discloses a training method for a palmprint recognition model, a palmprint recognition method, and related equipment. Compared with existing technologies, this invention acquires palmprint images collected from different devices and preprocesses the palmprint images according to the type of acquisition device. The preprocessed palmprint images are then input into a feature extraction network to extract image features. Based on the user and acquisition device of the palmprint image, positive and negative sample pairs are constructed for the image features to obtain a training sample set. The training sample set is then input into a pre-constructed contrastive learning model for training until a preset convergence condition is met, thus completing the model training and obtaining a palmprint recognition model. By using different feature processing methods for palmprint images collected from different devices and constructing positive and negative sample pairs for contrastive learning, the trained palmprint recognition model has stronger cross-device adaptability and improves the accuracy of cross-device palmprint recognition. Attached Figure Description

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0045] Figure 1 A flowchart illustrating the training method for a palmprint recognition model provided in an embodiment of the present invention;

[0046] Figure 2 A flowchart of step S300 in the training method of the palmprint recognition model provided in the embodiment of the present invention;

[0047] Figure 3 A flowchart of a palmprint recognition method provided in an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of the functional modules of the training device for the palmprint recognition model provided in an embodiment of the present invention;

[0049] Figure 5 This is a schematic diagram of the functional modules of the palmprint recognition device provided in an embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention is further described in detail below. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The embodiments of the invention are described below in conjunction with the accompanying drawings.

[0052] Palmprint recognition already has some similar products and applications in the financial sector. For example, financial institutions have begun using palmprint recognition technology as a means of identity verification and transaction authentication. In addition, some mobile payment platforms have also begun to introduce palmprint payment as a convenient payment method.

[0053] In palmprint payment and financial transaction scenarios, there may be situations where users register their palmprints on a mobile device, but the actual verification scenario is not on a mobile device. Device differences may lead to inconsistent palmprint image acquisition conditions. For example, factors such as lighting intensity, angle, and image quality between different devices may result in inconsistent palmprint image quality. Existing palmprint recognition technology has not been optimized for the problems that may arise from differences between different devices. This means that when performing palmprint recognition across devices, device differences may affect palmprint recognition, thus impacting the reliability of palmprint recognition.

[0054] To address the aforementioned problems, this invention proposes a training method for a palmprint recognition model and a palmprint recognition method. The system architecture of the palmprint recognition model training method or the palmprint recognition method provided in this embodiment may include a first terminal device, a second terminal device, a third terminal device, a network, and a server. The network serves as the medium for providing communication links between the first terminal device, the second terminal device, the third terminal device, and the server. The network may include various connection types, such as wired and / or wireless communication links, etc.

[0055] Users can use a first terminal device, a second terminal device, and a third terminal device to interact with the server via the network to receive or send messages, etc. Various communication client applications can be installed on the first terminal device, the second terminal device, and the third terminal device, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social media platform software, etc. (for example only).

[0056] The first terminal device, the second terminal device, and the third terminal device can be various electronic devices with a display screen and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0057] A server can be a server that provides various services, such as a backend management server that supports content viewed by users using a first terminal device, a second terminal device, or a third terminal device (this is just an example). The backend management server can analyze and process data such as received user requests and feed back the processing results (such as web pages, information, or data obtained or generated based on user requests) to the terminal devices. A server can also be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") in terms of management difficulty and weak business scalability. A server can also be a server for a distributed system or a server integrated with blockchain technology.

[0058] It should be noted that the training method or palmprint recognition method of the palmprint recognition model provided in the embodiments of the present invention can generally be executed by a first terminal device, a second terminal device, or a third terminal device. Accordingly, the training device or palmprint recognition device of the palmprint recognition model provided in the embodiments of the present invention can also be disposed in the first terminal device, the second terminal device, or the third terminal device.

[0059] Alternatively, the training method or palmprint recognition method of the palmprint recognition model provided in the embodiments of the present invention can generally be executed by a server. Correspondingly, the training device or palmprint recognition device of the palmprint recognition model provided in the embodiments of the present invention can generally be located in a server. The training method or palmprint recognition method of the palmprint recognition model provided in the embodiments of the present invention can also be executed by a server or server cluster that is different from the server and capable of communicating with the first terminal device, the second terminal device, the third terminal device, and / or the server. Correspondingly, the training device or palmprint recognition device of the palmprint recognition model provided in the embodiments of the present invention can also be located in a server or server cluster that is different from the server and capable of communicating with the first terminal device, the second terminal device, the third terminal device, and / or the server.

[0060] It should be understood that the number of terminal devices, networks, and servers listed above is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be used.

[0061] Please see Figure 1 , Figure 1 This is a flowchart of an embodiment of the training method for the palmprint recognition model provided by the present invention. Figure 1 As shown, the method specifically includes the following steps:

[0062] S100: Acquire palmprint images from different devices and perform corresponding preprocessing on the palmprint images based on the type of acquisition device.

[0063] In this embodiment, during the data collection and preprocessing stage, palm print images of multiple users are collected through different devices. For example, palm print images of the same user can be collected through mobile terminals, POS machines, ATMs, access control devices, smart locks, etc., thereby collecting a large-scale cross-device palm print image dataset to ensure diversity and representativeness.

[0064] This invention preprocesses palmprint images collected from multiple users across various devices. This preprocessing includes image normalization, cropping, and enhancement, such as adjusting image size, brightness and contrast, and performing histogram equalization. Preprocessing adjusts image quality to meet the requirements of palmprint recognition algorithms and reduces noise and artifacts, thereby improving the robustness of the palmprint recognition system under various environmental conditions. Furthermore, since device differences may lead to inconsistent palmprint image acquisition conditions (e.g., different devices may operate under different environmental conditions including lighting and background noise), this embodiment differentiates preprocessing based on the type of acquisition device, performing corresponding preprocessing on palmprint images acquired by different devices. This allows the preprocessing process to adapt to the characteristics of different device types, improving the cross-device adaptability and accuracy of palmprint recognition.

[0065] In one embodiment, the preprocessing of the palmprint image based on the type of acquisition device includes:

[0066] Confirm whether the device used to acquire the palmprint image is a mobile terminal;

[0067] When the acquisition device is a mobile terminal, the palmprint image is preprocessed according to the first preprocessing strategy;

[0068] When the acquisition device is not a mobile terminal, the palmprint image is preprocessed according to the second preprocessing strategy.

[0069] In this embodiment, when performing targeted preprocessing on the collected palmprint image dataset, the distinction is made based on whether the acquisition device is a mobile terminal. Mobile terminals can specifically include smartphones, laptops, tablets, etc., while non-mobile terminals can specifically include, for example, POS machines, ATMs, access control devices, smart locks, etc. If the palmprint image acquisition device is a mobile terminal, the palmprint image is preprocessed according to the first preprocessing strategy; otherwise, the palmprint image is preprocessed according to the second preprocessing strategy. Since images captured by mobile terminals may be affected by more environmental factors, such as unstable lighting and handheld device shake, while non-mobile terminals typically operate in more controlled environments, targeted preprocessing of palmprint images based on whether the acquisition device is a mobile terminal can adapt to the imaging conditions of different devices, thereby better adapting to the features in the image. Furthermore, different preprocessing strategies can ensure that image data from different types of terminals can be seamlessly integrated into a unified recognition system, thus improving the accuracy of cross-device palmprint recognition.

[0070] In one embodiment, preprocessing the palmprint image according to the first preprocessing strategy includes:

[0071] The size of the palm print image is adjusted to fit the target size of the mobile terminal;

[0072] The brightness and contrast of the palmprint image are adaptively adjusted.

[0073] The adaptively adjusted palmprint image is then subjected to noise reduction processing.

[0074] In this embodiment, for palmprint images captured by a mobile terminal, which typically has a small screen resolution, the size of the palmprint image is adjusted to fit the target size of the mobile terminal, such as 200x200 pixels. Then, the palmprint image undergoes adaptive brightness and contrast adjustment. Since the ambient lighting conditions of the mobile terminal may vary, adaptive adjustment of the palmprint image's brightness and contrast can be performed. For example, adjusting brightness and contrast by compressing the dynamic range ensures good visibility at different brightness levels. Alternatively, machine learning or deep learning models can be used to learn the optimal brightness and contrast adjustment strategy based on a large amount of image data to automatically adapt to different images. Another approach is to set a feedback loop to fine-tune the brightness and contrast parameters based on the adjusted image effect until a satisfactory result is achieved, thereby improving image clarity and recognizability. Finally, the adaptively adjusted palmprint image undergoes denoising processing. Since the mobile terminal's camera may introduce some noise, denoising algorithms such as mean filtering or median filtering can be used to reduce the impact of image noise on palmprint recognition.

[0075] In one embodiment, preprocessing the palmprint image according to the second preprocessing strategy includes:

[0076] The size of the palmprint image is adjusted to match the target size of the mobile terminal;

[0077] The palmprint image is subjected to histogram equalization and its color space is converted to a specified color space.

[0078] In this embodiment, for palmprint images collected from non-mobile terminals, the palmprint image is first adjusted to the same size as that of the mobile terminal, and histogram equalization is used to enhance the contrast of the image, making the palmprint texture more obvious and facilitating feature extraction and recognition. Then, according to the color space requirements of the device, the color space of the palmprint image is converted to a specified color space, such as from RGB color space to grayscale image or HSV color space, to facilitate subsequent processing and feature extraction, thereby improving the cross-device adaptability of palmprint recognition.

[0079] S200. Input the preprocessed palmprint image into a feature extraction network to extract the image features of the palmprint image.

[0080] In this embodiment, a deep neural network, such as MobileNetV3 or ResNet, is used as the feature extractor. MobileNetV3 is a lightweight deep learning architecture that optimizes model size and computational efficiency, making it suitable for running on resource-constrained devices. ResNet, a residual network, introduces residual modules, allowing signals in the network to skip one or more layers, helping to solve the gradient vanishing problem in deep networks and extracting features from different levels of the image. The feature extraction network can be flexibly configured according to requirements. Specifically, the weight parameters of the feature extraction network can use pre-trained network weights, or the network weights can be initialized first and then trained on a large-scale public image dataset to obtain better feature extraction capabilities. The trained feature extraction network extracts features from the pre-processed palmprint image. The palmprint image is input into the feature extraction network to obtain the corresponding feature vector. These image features will be used in subsequent contrastive learning tasks to achieve reliable and accurate palmprint recognition.

[0081] S300. Based on the user and acquisition device to which the palm print image belongs, construct positive and negative sample pairs for the image features to obtain a training sample set.

[0082] In this embodiment, based on the image features extracted from the palmprint image dataset, positive and negative sample pairs are constructed for all image features according to the user and acquisition device of the palmprint image. This results in a training sample set consisting of multiple sets of positive and negative sample pairs, which are used for subsequent comparative learning of the model. Positive sample pairs refer to data pairs that should be considered similar or belong to the same category in the feature space. In comparative learning, the model is trained to increase the similarity of these data pairs. Negative sample pairs refer to data pairs that should be considered dissimilar or belong to different categories in the feature space. By constructing positive and negative sample pairs, the model is guided to learn during training. This allows the model to learn the similarity of palmprint images acquired by the user on different devices, enabling the model to adapt to different acquisition environments and thus improve cross-device palmprint recognition.

[0083] In one embodiment, such as Figure 2 As shown, based on the user and acquisition device to which the palmprint image belongs, the image features are used to construct a training sample set by pairing positive and negative samples, including:

[0084] Positive sample pairs are constructed by using image features collected from the same user on different devices as positive samples.

[0085] Select negative samples that are different from the positive samples from image features collected by other users on any device, and construct negative sample pairs with the positive samples;

[0086] The positive and negative sample pairs are combined to form a positive-negative sample pair, thus obtaining the training sample set.

[0087] In this embodiment, when constructing the training sample set, feature alignment can be performed on image features acquired from different devices to reduce differences between devices. For example, feature alignment algorithms, including Procrustes analysis or registration algorithms, can be used to align image features from different devices, making them more consistent in the feature space. Then, positive and negative sample pairs are constructed based on the user and acquisition device of the palmprint image. For image features acquired from the same user on different devices, since they belong to the same user and represent the same palmprint features, they are considered positive samples. One or more pairs are randomly selected from all positive samples as positive sample pairs. Simultaneously, negative samples different from the positive samples are selected from image features acquired from other users on any device or randomly sampled images, and these are used to construct negative sample pairs with any positive sample. These negative sample pairs represent different palmprint features; that is, the negative sample pairs are palmprint image feature pairs from different users, possibly from the same type of acquisition device or different types of acquisition devices. Based on the generated positive and negative sample pairs, a positive-negative sample pair is constructed to obtain a training sample set for contrastive learning. In contrastive learning, the model is trained to increase the similarity of positive sample pairs and decrease the similarity of negative sample pairs. This helps the palmprint recognition system to better distinguish and recognize the palmprints of different people even when there are changes in the palmprint image (such as angle, lighting, blur, etc.).

[0088] Preferably, to increase the diversity and generalization ability of the training samples, sample amplification operations can be performed on the acquired image features. For example, operations such as image rotation, translation, scaling, or adding noise can be applied to generate multiple deformed image features as training samples, thereby improving the diversity and richness of the training sample set and further improving the accuracy of palmprint recognition.

[0089] S400. Input the training sample set into the pre-built contrastive learning model for training until the preset convergence condition is met, then the model training is completed and a palmprint recognition model is obtained.

[0090] In this embodiment, a contrastive learning model, such as a Siamese network or a Triplet network, is pre-built. Contrastive learning models learn the differences between different data points by comparing them. This type of model is commonly used in unsupervised or semi-supervised learning tasks, especially when there is a large amount of unlabeled data in the dataset. Since the goal of contrastive learning models is to identify and distinguish different patterns or categories, even if these categories do not have explicit labels, a key characteristic of contrastive learning models is that they emphasize the relative relationships between data points rather than absolute values. This allows contrastive learning models to effectively learn the similarity between palmprint images of the same user collected from different devices when used for palmprint recognition tasks, thus improving cross-device adaptability. Furthermore, based on the contrastive learning approach, it is not necessary to store the original palmprint images; only a relatively small feature vector needs to be stored, further enhancing privacy protection capabilities.

[0091] The constructed training sample set is input into the contrastive learning model for training. Specifically, the contrastive learning loss function, such as Siamese network loss or triplet loss, can be used. During training, for each sample pair, the loss is calculated by comparing the similarity or difference between the two features. The loss function is optimized to make feature pairs from the same user closer and feature pairs from different users more separated. In other words, the contrastive learning model optimizes the value of the loss function to maximize the similarity between positive sample pairs and minimize the similarity between negative sample pairs. This makes palmprint images from the same person but different devices (positive sample pairs) closer in the feature space, while palmprint images from different people (negative sample pairs) farther apart. Through contrastive learning, the trained palmprint recognition model can better distinguish and recognize the palmprints of different people when multiple acquisition devices are present, improving the cross-device adaptability and reliability of palmprint recognition.

[0092] It should be noted that there is no necessary order between the above steps. Those skilled in the art will understand from the description of the embodiments of the present invention that the above steps may have different execution orders in different embodiments, that is, they may be executed in parallel or in turn, etc.

[0093] Another embodiment of the present invention provides a palmprint recognition method, such as... Figure 3 As shown, the method includes the following steps:

[0094] S500: Acquire the palmprint image to be identified and extract the image features of the palmprint image to be identified;

[0095] S600 inputs the image features of the palmprint image to be identified into a pre-trained palmprint recognition model, compares and identifies the palmprints with the palmprint database of registered users, and outputs the palmprint recognition result of the palmprint image to be identified.

[0096] The pre-trained palmprint recognition model is obtained using the training method described above.

[0097] In this embodiment, after training the palmprint recognition model using the training method described above, cross-device palmprint recognition can be achieved in various application scenarios. For example, a user's palmprint image can be collected on a mobile terminal, and palmprint image recognition and verification can be performed on a POS machine. In this case, the palmprint image collected on the POS machine is used as the palmprint image to be recognized, and a feature extraction network is used to extract the image features of the palmprint image to be recognized. The image features of the palmprint image to be recognized are input into the pre-trained palmprint recognition model. The palmprint recognition model compares the image features with the image features in the palmprint database of registered users. The model can determine whether the palmprint image to be recognized is successfully matched by calculating the similarity between image features, such as cosine similarity or Euclidean distance. Since the palmprint recognition model adopts different feature processing methods for images collected from different devices during the training phase, and obtains the model through comparative learning, it has strong cross-device adaptability. Even if there are differences in the palmprint images collected from different devices, the model can learn more robust feature representations, improving the cross-device recognition accuracy.

[0098] Another embodiment of the present invention provides a training device for a palmprint recognition model, such as... Figure 4 As shown, the training device 1 includes:

[0099] The data collection and processing module 11 acquires palmprint images collected from different devices and performs corresponding preprocessing on the palmprint images based on the type of acquisition device.

[0100] Feature extraction module 12 is used to input the preprocessed palmprint image into the feature extraction network to extract the image features of the palmprint image;

[0101] The sample construction module 13 is used to construct positive and negative sample pairs for the image features based on the user and acquisition device to which the palm print image belongs, so as to obtain a training sample set;

[0102] The model training module 14 is used to input the training sample set into the pre-built contrastive learning model for training until the preset convergence condition is met, and then the model training is completed to obtain the palmprint recognition model.

[0103] The module referred to in this invention is a series of computer program instruction segments that can perform specific functions. It is more suitable than a program for describing the training and execution process of the palmprint recognition model. For the specific implementation of each module, please refer to the corresponding method embodiments above, which will not be repeated here.

[0104] In one embodiment, the data collection and processing module 11 includes:

[0105] A type confirmation unit is used to confirm whether the palmprint image acquisition device is a mobile terminal;

[0106] The first preprocessing unit is used to preprocess the palmprint image according to the first preprocessing strategy when the acquisition device is a mobile terminal.

[0107] The second preprocessing unit is used to preprocess the palmprint image according to the second preprocessing strategy when the acquisition device is not a mobile terminal.

[0108] In one embodiment, the first preprocessing unit includes:

[0109] The first size adjustment unit is used to adjust the size of the palm print image to fit the target size of the mobile terminal;

[0110] An adaptive adjustment unit is used to adaptively adjust the brightness and contrast of the palmprint image;

[0111] A denoising unit is used to denoise the adaptively adjusted palmprint image.

[0112] In one embodiment, the second preprocessing unit includes:

[0113] The second size adjustment unit is used to adjust the size of the palm print image to a target size consistent with the mobile terminal;

[0114] The equalization and conversion unit is used to perform histogram equalization processing on the palm print image and convert the color space of the palm print image to a specified color space.

[0115] In one embodiment, the sample construction module 13 includes:

[0116] The positive sample pair construction unit is used to construct positive sample pairs by taking image features collected from the same user on different devices as positive samples.

[0117] The negative sample pair construction unit is used to select negative samples that are different from the positive samples from image features collected by other users on any device, and construct negative sample pairs with the positive samples;

[0118] The sample set construction unit is used to construct positive and negative sample pairs from the positive and negative sample pairs to obtain the training sample set.

[0119] Another embodiment of the present invention provides a palmprint recognition device, such as... Figure 5 As shown, device 2 includes:

[0120] Image acquisition module 21 is used to acquire a palm print image to be identified and extract the image features of the palm print image to be identified;

[0121] The recognition module 22 is used to input the image features of the palm print image to be recognized into a pre-trained palm print recognition model, compare and recognize the palm print with the palm print database of registered users, and output the palm print recognition result of the palm print image to be recognized.

[0122] The pre-trained palmprint recognition model was obtained using the training method described above.

[0123] The image acquisition module 21 and the recognition module 22 are connected. The module referred to in this invention is a series of computer program instruction segments that can perform specific functions. It is more suitable than a program for describing the palmprint recognition execution process. For the specific implementation of each module, please refer to the corresponding method embodiments above, which will not be repeated here.

[0124] Another embodiment of the present invention provides an electronic device, such as... Figure 6 As shown, the electronic device 10 includes:

[0125] One or more processors 110 and memory 120, Figure 6 The following description uses a processor 110 as an example. The processor 110 and the memory 120 can be connected via a bus or other means. Figure 6 Taking the example of a connection between China and Israel via a bus.

[0126] Processor 110 is used to perform various control logics of electronic device 10. It can be a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), microcontroller, ARM (Acorn RISC Machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components. Furthermore, processor 110 can also be any conventional processor, microprocessor, or state machine. Processor 110 can also be implemented as a combination of computing devices, such as a combination of DSP and microprocessor, multiple microprocessors, one or more microprocessors combined with DSP and / or any other such configuration.

[0127] The memory 120, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions corresponding to the palmprint recognition model training method in the embodiments of the present invention. The processor 110 executes various functional applications and data processing of the electronic device 10 by running the non-volatile software programs, instructions, and units stored in the memory 120, that is, implementing the palmprint recognition model training method or palmprint recognition method in the above method embodiments.

[0128] The memory 120 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device 10. Furthermore, the memory 120 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 120 may optionally include memory remotely located relative to the processor 110, and these remote memories may be connected to the electronic device 10 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0129] One or more units are stored in memory 120. When executed by one or more processors 110, they perform the training method of the palmprint recognition model in any of the above method embodiments, for example, performing the following steps:

[0130] Acquire palmprint images from different devices and perform corresponding preprocessing on the palmprint images based on the type of acquisition device;

[0131] The preprocessed palmprint image is input into a feature extraction network to extract the image features of the palmprint image;

[0132] Based on the user and acquisition device of the palmprint image, positive and negative sample pairs are constructed from the image features to obtain a training sample set;

[0133] The training sample set is input into a pre-built contrastive learning model for training until a preset convergence condition is met, at which point the model training is complete, and a palmprint recognition model is obtained.

[0134] In one embodiment, the preprocessing of the palmprint image based on the type of acquisition device includes:

[0135] Confirm whether the device used to acquire the palmprint image is a mobile terminal;

[0136] When the acquisition device is a mobile terminal, the palmprint image is preprocessed according to the first preprocessing strategy;

[0137] When the acquisition device is not a mobile terminal, the palmprint image is preprocessed according to the second preprocessing strategy.

[0138] In one embodiment, preprocessing the palmprint image according to the first preprocessing strategy includes:

[0139] The size of the palm print image is adjusted to fit the target size of the mobile terminal;

[0140] The brightness and contrast of the palmprint image are adaptively adjusted.

[0141] The adaptively adjusted palmprint image is then subjected to noise reduction processing.

[0142] In one embodiment, preprocessing the palmprint image according to the second preprocessing strategy includes:

[0143] The size of the palmprint image is adjusted to match the target size of the mobile terminal;

[0144] The palmprint image is subjected to histogram equalization and its color space is converted to a specified color space.

[0145] In one embodiment, the step of constructing a training sample set by pairing positive and negative samples based on the user and acquisition device of the palmprint image includes:

[0146] Positive sample pairs are constructed by using image features collected from the same user on different devices as positive samples.

[0147] Select negative samples that are different from the positive samples from image features collected by other users on any device, and construct negative sample pairs with the positive samples;

[0148] The positive and negative sample pairs are combined to form a positive-negative sample pair, thus obtaining the training sample set.

[0149] Alternatively, the palmprint recognition method in any of the above method embodiments can be executed, for example, by performing the following steps:

[0150] Acquire a palmprint image to be identified, and extract the image features of the palmprint image to be identified;

[0151] The image features of the palmprint image to be identified are input into a pre-trained palmprint recognition model, and the palmprint is compared and recognized with the palmprint database of registered users. The palmprint recognition result of the palmprint image to be identified is then output.

[0152] The pre-trained palmprint recognition model was obtained using the training method described above.

[0153] This invention also provides a non-volatile computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are executed by one or more processors, they perform the training method for the palmprint recognition model in any of the above method embodiments, for example, performing the following steps:

[0154] Acquire palmprint images from different devices and perform corresponding preprocessing on the palmprint images based on the type of acquisition device;

[0155] The preprocessed palmprint image is input into a feature extraction network to extract the image features of the palmprint image;

[0156] Based on the user and acquisition device of the palmprint image, positive and negative sample pairs are constructed from the image features to obtain a training sample set;

[0157] The training sample set is input into a pre-built contrastive learning model for training until a preset convergence condition is met, at which point the model training is complete, and a palmprint recognition model is obtained.

[0158] In one embodiment, the preprocessing of the palmprint image based on the type of acquisition device includes:

[0159] Confirm whether the device used to acquire the palmprint image is a mobile terminal;

[0160] When the acquisition device is a mobile terminal, the palmprint image is preprocessed according to the first preprocessing strategy;

[0161] When the acquisition device is not a mobile terminal, the palmprint image is preprocessed according to the second preprocessing strategy.

[0162] In one embodiment, preprocessing the palmprint image according to the first preprocessing strategy includes:

[0163] The size of the palm print image is adjusted to fit the target size of the mobile terminal;

[0164] The brightness and contrast of the palmprint image are adaptively adjusted.

[0165] The adaptively adjusted palmprint image is then subjected to noise reduction processing.

[0166] In one embodiment, preprocessing the palmprint image according to the second preprocessing strategy includes:

[0167] The size of the palmprint image is adjusted to match the target size of the mobile terminal;

[0168] The palmprint image is subjected to histogram equalization and its color space is converted to a specified color space.

[0169] In one embodiment, the step of constructing a training sample set by pairing positive and negative samples based on the user and acquisition device of the palmprint image includes:

[0170] Positive sample pairs are constructed by using image features collected from the same user on different devices as positive samples.

[0171] Select negative samples that are different from the positive samples from image features collected by other users on any device, and construct negative sample pairs with the positive samples;

[0172] The positive and negative sample pairs are combined to form a positive-negative sample pair, thus obtaining the training sample set.

[0173] Alternatively, the palmprint recognition method in any of the above method embodiments can be executed, for example, by performing the following steps:

[0174] Acquire a palmprint image to be identified, and extract the image features of the palmprint image to be identified;

[0175] The image features of the palmprint image to be identified are input into a pre-trained palmprint recognition model, and the palmprint is compared and recognized with the palmprint database of registered users. The palmprint recognition result of the palmprint image to be identified is then output.

[0176] The pre-trained palmprint recognition model was obtained using the training method described above.

[0177] As examples, non-volatile storage media can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) as external cache memory. By way of illustration and not limitation, RAM can be obtained in many forms such as synchronous RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The memory components or memories disclosed in the operating environment described herein are intended to include one or more of these and / or any other suitable types of memory.

[0178] In summary, the palmprint recognition model training method, palmprint recognition method, and related equipment disclosed in this invention include the following training method: acquiring palmprint images collected from different devices, and performing corresponding preprocessing on the palmprint images based on the type of acquisition device; inputting the preprocessed palmprint images into a feature extraction network to extract image features from the palmprint images; constructing positive and negative sample pairs for the image features based on the user and acquisition device of the palmprint images to obtain a training sample set; and inputting the training sample set into a pre-constructed contrastive learning model for training until a preset convergence condition is met, thus completing the model training and obtaining the palmprint recognition model. By using different feature processing methods on palmprint images collected from different devices and constructing positive and negative sample pairs for contrastive learning, the trained palmprint recognition model has stronger cross-device adaptability and improves the accuracy of cross-device palmprint recognition.

[0179] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The computer program can be stored in a non-volatile, computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The storage medium can be a memory, magnetic disk, floppy disk, flash memory, optical storage, etc.

[0180] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A training method for a palmprint recognition model, characterized in that, include: Acquire palmprint images from different devices and perform corresponding preprocessing on the palmprint images based on the type of acquisition device; The preprocessed palmprint image is input into a feature extraction network to extract the image features of the palmprint image; Feature alignment processing is performed on image features collected from different devices. Based on the user and the collection device of the palm print image, positive and negative sample pairs are constructed for the image features to obtain a training sample set. The training sample set is input into a pre-built contrastive learning model for training until a preset convergence condition is met, at which point the model training is complete and a palmprint recognition model is obtained. The preprocessing of the palmprint image based on the type of acquisition device includes: Confirm whether the device used to acquire the palmprint image is a mobile terminal; When the acquisition device is a mobile terminal, the palmprint image is preprocessed according to the first preprocessing strategy; When the acquisition device is not a mobile terminal, the palmprint image is preprocessed according to the second preprocessing strategy; The preprocessing of the palmprint image according to the first preprocessing strategy includes: The size of the palm print image is adjusted to fit the target size of the mobile terminal; The palmprint image is adaptively adjusted for brightness and contrast by compressing the dynamic range, using a machine learning model, or a deep learning model. The adaptively adjusted palmprint image is then subjected to denoising processing. The preprocessing of the palmprint image according to the second preprocessing strategy includes: The size of the palmprint image is adjusted to match the target size of the mobile terminal; The palmprint image is subjected to histogram equalization and its color space is converted to a specified color space.

2. The training method for the palmprint recognition model according to claim 1, characterized in that, Based on the user and acquisition device associated with the palmprint image, the image features are used to construct positive and negative sample pairs to obtain a training sample set, including: Positive sample pairs are constructed by using image features collected from the same user on different devices as positive samples. Select negative samples that are different from the positive samples from image features collected by other users on any device, and construct negative sample pairs with the positive samples; The positive and negative sample pairs are combined to form a positive-negative sample pair, thus obtaining the training sample set.

3. A palmprint recognition method, characterized in that, include: Acquire a palmprint image to be identified, and extract the image features of the palmprint image to be identified; The image features of the palmprint image to be identified are input into a pre-trained palmprint recognition model, and the palmprint is compared and recognized with the palmprint database of registered users. The palmprint recognition result of the palmprint image to be identified is then output. The pre-trained palmprint recognition model is obtained using the training method described in any one of claims 1-2.

4. A training device for a palmprint recognition model, characterized in that, include: The data collection and processing module acquires palmprint images collected from different devices and performs corresponding preprocessing on the palmprint images based on the type of acquisition device. The feature extraction module is used to input the preprocessed palmprint image into the feature extraction network to extract the image features of the palmprint image; The sample construction module is used to perform feature alignment processing on image features collected from different devices. Based on the user and the collection device of the palmprint image, positive and negative sample pairs are constructed on the image features to obtain a training sample set. The model training module is used to input the training sample set into the pre-built contrastive learning model for training until the preset convergence condition is met, then the model training is completed and a palmprint recognition model is obtained. The data collection and processing module includes: A type confirmation unit is used to confirm whether the palmprint image acquisition device is a mobile terminal; The first preprocessing unit is used to preprocess the palmprint image according to the first preprocessing strategy when the acquisition device is a mobile terminal. The second preprocessing unit is used to preprocess the palmprint image according to the second preprocessing strategy when the acquisition device is not a mobile terminal. The first preprocessing unit includes: The first size adjustment unit is used to adjust the size of the palm print image to fit the target size of the mobile terminal; An adaptive adjustment unit is used to adaptively adjust the brightness and contrast of the palmprint image by compressing the dynamic range, using a machine learning model, or a deep learning model. A denoising unit is used to denoise the adaptively adjusted palmprint image; The second preprocessing unit includes: The second size adjustment unit is used to adjust the size of the palm print image to a target size consistent with the mobile terminal; The equalization and conversion unit is used to perform histogram equalization processing on the palm print image and convert the color space of the palm print image to a specified color space.

5. A palmprint recognition device, characterized in that, include: The image acquisition module is used to acquire a palmprint image to be identified and extract the image features of the palmprint image to be identified; The recognition module is used to input the image features of the palmprint image to be recognized into a pre-trained palmprint recognition model, compare and recognize the palmprint with the palmprint database of registered users, and output the palmprint recognition result of the palmprint image to be recognized. The pre-trained palmprint recognition model is obtained using the training method described in any one of claims 1-2.

6. An electronic device, characterized in that, Includes at least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the training method of the palmprint recognition model according to any one of claims 1-2, or to perform the palmprint recognition method according to claim 3.

7. A non-volatile computer-readable storage medium, characterized in that, The non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by one or more processors, cause the one or more processors to perform the training method of the palmprint recognition model according to any one of claims 1-2, or to perform the palmprint recognition method according to claim 3.