Finger identification method and device based on finger vein, electronic equipment and storage medium

Abstract

This application provides a finger recognition method, apparatus, electronic device, and storage medium based on finger veins. The method includes: acquiring a current finger vein image by performing image acquisition on a target finger vein of a target object; querying historical finger vein images that match the current finger vein image; predicting the blood vessel direction in a blurred region of the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred region, wherein the blurred region is a region in the current finger vein image where the blood vessel direction cannot be determined; and determining the target finger as a specified finger if the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, wherein the historical finger vein sub-image is a sub-image in the historical finger vein image corresponding to the blurred region. This application overcomes the technical problem in related technologies where finger injury during user use leads to finger vein recognition failure.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, and in particular to a finger recognition method and apparatus, electronic device and storage medium based on finger veins. Background Technology

[0002] Finger vein recognition technology is an emerging biometric technology. Due to its strong anti-counterfeiting properties and low cost, finger vein recognition is more secure than fingerprint recognition because it is uncopyable. Furthermore, it can recognize hands regardless of whether they are too wet or too dry, and can quickly identify the hands of the elderly and children.

[0003] In related technologies, pre-recorded and stored images of the user's finger veins are used to verify the characteristics of the finger veins when opening a door. However, if the user injures their finger during use, the finger vein recognition will fail.

[0004] Therefore, there is a problem in the relevant technology that if the user injures their finger during use, it will cause the finger vein recognition to fail. Summary of the Invention

[0005] This application provides a finger recognition method and apparatus, electronic device and storage medium based on finger veins, to at least solve the problem in the related art that finger vein recognition will fail when the user injures their finger during use.

[0006] According to one aspect of the embodiments of this application, a finger recognition method based on finger veins is provided, including:

[0007] The current finger vein image is obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with a wound.

[0008] The query retrieves historical finger vein images that match the current finger vein image. These historical finger vein images are obtained by capturing images of the specified finger veins when the specified finger has no wounds. The specified finger is a finger whose finger vein information has been pre-entered.

[0009] The blood vessel course in the blurred region of the current finger vein image is predicted to obtain a predicted finger vein sub-image corresponding to the blurred region, wherein the blurred region is the region in the current finger vein image where the blood vessel course cannot be determined;

[0010] If the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, the target finger is determined to be the designated finger, wherein the historical finger vein sub-image is the sub-image in the historical finger vein image corresponding to the blurred region.

[0011] Optionally, as described above, the step of querying to obtain historical finger vein images that match the current finger vein image includes:

[0012] Obtain all candidate finger vein images;

[0013] Determine the matching degree between all feature points of each candidate finger vein image and all feature points of the current finger vein image;

[0014] Based on the matching degree, the historical finger vein image is determined from all the candidate finger vein images, wherein the matching degree between the historical finger vein image and the current finger vein image is greater than or equal to a preset matching degree threshold.

[0015] Optionally, as described above, predicting the blood vessel course in the blurred region of the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred region includes:

[0016] Starting from the edge of the blurred region, the blurred region is differentiated to obtain a series of consecutive differentiated images;

[0017] Each of the differential images is sequentially input into a preset target deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image;

[0018] The predicted finger vein sub-image is obtained based on all the predicted differential images.

[0019] Optionally, as described above, the step of sequentially inputting each of the differential images into a preset deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image includes:

[0020] Determine the vanishing point of the blood vessel path corresponding to the differential image, wherein, when the differential image is the edge of the blurred region, the vanishing point of the blood vessel path corresponding to the differential image is the vanishing point of the blood vessel path in the clear region; when the differential image is no longer the edge of the blurred region, the vanishing point of the blood vessel path corresponding to the differential image is the vanishing point of the blood vessel path in the predicted differential image obtained from the previous differential image.

[0021] The vanishing point of the blood vessel path and the differential image are input into the target deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in the differential image.

[0022] Optionally, as described above, before sequentially inputting each of the differential images into a preset deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image, the method further includes:

[0023] Obtain a training dataset and a validation dataset, wherein the training dataset includes multiple sets of training data, and the validation dataset includes multiple sets of training data, each set of training data including corresponding blood vessel path vanishing points and training differential images;

[0024] The deep learning algorithm to be trained is trained using the training dataset to obtain the trained deep learning algorithm.

[0025] If the trained deep learning algorithm is validated using the validation dataset and it is determined that the trained deep learning algorithm meets the preset accuracy requirements, then the trained deep learning algorithm is identified as the target deep learning algorithm.

[0026] Optionally, as described above, determining the target finger as the designated finger when the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold includes:

[0027] Determine the first vessel orientation information of the historical finger vein sub-image and the second vessel orientation information of the predicted finger vein sub-image;

[0028] If the preset similarity threshold is met between the first blood vessel direction information and the second blood vessel direction information, the target finger is determined to be the designated finger.

[0029] Optionally, as described above, after determining that the target finger is the designated finger, the method further includes:

[0030] When the current finger vein image is used to unlock the lock, the motor of the lock is controlled to operate in order to unlock the lock.

[0031] According to another aspect of the embodiments of this application, a finger recognition device based on finger veins is also provided, comprising:

[0032] The acquisition module is used to acquire the current finger vein image obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with a wound.

[0033] The query module is used to query and obtain historical finger vein images that match the current finger vein image. The historical finger vein image is a finger vein image obtained by acquiring the image of the specified finger vein when the specified finger has no wound. The specified finger is a finger whose finger vein information has been pre-entered.

[0034] The prediction module is used to predict the blood vessel direction in the blurred area of ​​the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred area, wherein the blurred area is the area in the current finger vein image where the blood vessel direction cannot be determined;

[0035] The determination module is used to determine the target finger as the specified finger when the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, wherein the historical finger vein sub-image is a sub-image in the historical finger vein image corresponding to the blurred region.

[0036] According to another aspect of the embodiments of this application, an electronic device is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; wherein the memory is used to store a computer program; and the processor is used to execute the method steps of any of the above embodiments by running the computer program stored in the memory.

[0037] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to execute the method steps of any of the above embodiments when running.

[0038] In this embodiment, a finger recognition method and device based on finger veins, an electronic device, and a storage medium are employed. The method involves acquiring a current finger vein image obtained by image acquisition of the target finger vein of a target object, wherein the target finger vein is the finger vein of a target finger with a current wound; querying historical finger vein images that match the current finger vein image, wherein the historical finger vein image is a finger vein image obtained by image acquisition of a specified finger vein when the specified finger does not have a wound, and the specified finger is a finger whose finger vein information has been pre-entered; predicting the blood vessel direction in a blurred region in the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred region, wherein the blurred region is a region in the current finger vein image where the blood vessel direction cannot be determined; and determining the target finger as the specified finger if the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, wherein the historical finger vein sub-image is a sub-image in the historical finger vein image corresponding to the blurred region. By predicting the blood vessel course in the blurred area of ​​the current finger vein image, a predicted finger vein sub-image corresponding to the blurred area is obtained. Furthermore, based on the predicted finger vein sub-image, it is determined whether the target finger is the specified finger. This can achieve the goal of determining whether the current finger vein image is a historical finger vein image even when the target finger is injured, i.e., when there is a blurred area in the current finger vein image. In this way, it can be determined whether the target finger is the specified finger, thus overcoming the technical problem in related technologies where finger vein recognition fails when the user injures their finger during use. Attached Figure Description

[0039] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic flowchart of an optional finger recognition method based on finger veins according to an embodiment of this application;

[0042] Figure 2 This is a flowchart illustrating an optional finger recognition method based on finger veins according to an application example of this application;

[0043] Figure 3This is a structural block diagram of an optional finger recognition device based on finger veins according to an embodiment of this application;

[0044] Figure 4 This is a structural block diagram of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] According to one aspect of the embodiments of this application, a finger recognition method based on finger veins is provided. Optionally, in this embodiment, the above-described finger recognition method based on finger veins can be applied to a hardware environment consisting of a terminal and a server. The server is connected to the terminal via a network and can be used to provide services (such as advertising push services, application services, etc.) to the terminal or clients installed on the terminal. A database can be set up on the server or independently of the server to provide data storage services to the server.

[0048] The aforementioned network may include, but is not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network, metropolitan area network, local area network. The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth. The terminal is not limited to PC, mobile phone, tablet computer, etc.

[0049] The finger recognition method based on finger veins in this application can be executed by a server, a terminal, or both. Alternatively, the finger recognition method based on finger veins in this application can be executed by a client installed on the terminal.

[0050] Taking the finger recognition method based on finger veins in this embodiment as an example, which is executed by a terminal, Figure 1 A finger recognition method based on finger veins, provided in this application embodiment, includes the following steps:

[0051] Step S101: Obtain the current finger vein image obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with the current wound.

[0052] The finger vein-based finger recognition method in this embodiment can be applied to scenarios requiring identity authentication through finger vein recognition, such as unlocking doors, clocking in, or other identity authentication scenarios. This embodiment uses the unlocking of a smart door lock as an example to illustrate the above-described finger vein-based finger recognition method. For other types of scenarios, the above-described finger vein-based finger recognition method is equally applicable, provided there is no contradiction.

[0053] Optionally, a current finger vein image of the target finger vein of the target object can be acquired using a device for acquiring finger vein images.

[0054] The target audience can be the user who needs to undergo finger vein recognition.

[0055] The target digital vein can be the digital vein of the target finger where there is currently an injury.

[0056] The target finger can be a finger that the target object has pre-selected for recording digital vein images, in the absence of any wounds.

[0057] Step S102: Query and obtain historical finger vein images that match the current finger vein image. The historical finger vein image is a finger vein image obtained by acquiring the image of the specified finger vein when there is no wound on the specified finger. The specified finger is the finger whose finger vein information has been pre-entered.

[0058] Optionally, after obtaining the current finger vein image, the smart lock can query the MCU to obtain historical finger vein images that match the current finger vein image.

[0059] Because the target finger has a wound, the wound or blood will cause some areas in the current finger vein image to be unable to accurately identify the path of the finger vein.

[0060] In this case, even if the target finger has been pre-recorded with the corresponding candidate finger vein image and stored in the MCU, it is impossible to obtain a completely consistent candidate finger vein image.

[0061] Therefore, the historical finger vein image can be determined by identifying the one with the highest matching degree to the current finger vein image among all candidate finger vein images.

[0062] As an optional embodiment, retrieving historical finger vein images that match the current finger vein image includes the following steps:

[0063] Step S201: Obtain all candidate finger vein images;

[0064] Step S202: Determine the matching degree between all feature points of each candidate finger vein image and all feature points of the current finger vein image in all candidate finger vein images;

[0065] Step S203: Based on the matching degree, determine the historical finger vein image from all candidate finger vein images, wherein the matching degree between the historical finger vein image and the current finger vein image is greater than or equal to a preset matching degree threshold.

[0066] All pre-recorded candidate finger vein images can be retrieved from the MCU and stored in the MCU.

[0067] Then, perform the following operations on each candidate finger vein image: extract all feature points in the candidate finger vein image and extract all feature points in the current finger vein image; then compare all feature points in the candidate finger vein image with the current finger vein image to obtain the matching degree between the candidate finger vein image and the current finger vein image.

[0068] Therefore, each candidate finger vein image has a matching degree. Based on this matching degree, historical finger vein images with a matching degree greater than or equal to a preset matching degree threshold can be identified among all candidate finger vein images.

[0069] Optionally, the preset matching threshold can be a threshold obtained based on empirical data, such as the maximum similarity that different finger veins can achieve under normal circumstances; in this embodiment, the preset matching threshold can be 80%.

[0070] Furthermore, after identifying historical finger vein images, the target ID corresponding to those historical finger vein images can also be determined, so that the identity information of the target object can be determined based on the target ID later.

[0071] Step S103: Predict the blood vessel direction in the blurred area of ​​the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred area, wherein the blurred area is the area in the current finger vein image where the blood vessel direction cannot be determined.

[0072] After obtaining the current finger vein image, because the target finger has a wound, there will be some areas in the image where the path of the finger vein cannot be accurately identified due to the wound or blood. Therefore, there will be blurred areas in the current finger vein image, which are the areas where the direction of the blood vessels cannot be determined.

[0073] The blurred region in the current finger vein image may include one or more regions. After obtaining the blurred region, the direction of blood vessels in the blurred region can be predicted based on the direction of blood vessels in other regions (i.e., regions where the direction of blood vessels can be determined) connected to the blurred region, thereby obtaining a predicted finger vein sub-image corresponding to the blurred region.

[0074] The predicted finger vein sub-image is a portion of the image corresponding to a blurred region. Furthermore, when there are multiple blurred regions, each blurred region has a corresponding predicted finger vein sub-image.

[0075] Step S104: If the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets the preset similarity threshold, the target finger is determined as the specified finger, wherein the historical finger vein sub-image is the sub-image in the historical finger vein image corresponding to the blurred region.

[0076] After obtaining the predicted finger vein sub-image, it is possible to further determine whether the target finger is the specified finger based on the predicted finger vein sub-image.

[0077] Optionally, a historical finger vein sub-image corresponding to the blurred area can be identified from the historical finger vein images.

[0078] Then, the similarity between the historical finger vein sub-image and the predicted finger vein sub-image is determined. If the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets the preset similarity threshold, the target finger is determined as the specified finger.

[0079] Optionally, the preset similarity threshold can be 90%. In addition, other thresholds can be selected according to actual debugging, which will not be listed here.

[0080] The method in this embodiment predicts the blood vessel course in the blurred area of ​​the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred area. Based on the predicted finger vein sub-image, it further determines whether the target finger is the designated finger. This method can still determine whether the current finger vein image is a historical finger vein image even when the target finger is injured, i.e., when there is a blurred area in the current finger vein image. In this way, it can determine whether the target finger is the designated finger, thereby overcoming the technical problem in related technologies that finger vein recognition will fail when the user injures their finger during use.

[0081] As an optional embodiment, step S103 predicts the blood vessel course in the blurred region of the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred region, including the following steps:

[0082] Step S301: Starting from the edge of the blurred region, differentiate the blurred region to obtain multiple consecutive differentiated images;

[0083] After obtaining the blurred region, the edge of the blurred region can be used as the starting point, and the blurred region can be differentiated according to the direction of the blood vessel path (e.g., the direction of the blood vessel path in other regions connected to the blurred region) to obtain a series of consecutive differentiated images. Generally, there are no identical parts between the multiple differentiated images, and the blurred region can be restored after the multiple differentiated images are stitched together according to the direction of the blood vessel path.

[0084] Step S302: Input each differential image into the preset target deep learning algorithm in sequence to obtain the predicted differential image after predicting the blood vessel path in each differential image;

[0085] After obtaining each differential image, the differential images can be sequentially input into a preset target deep learning algorithm according to the blood vessel path. The target deep learning algorithm then predicts the blood vessel path in each differential image to obtain a predicted differential image. That is, each differential image has a corresponding predicted differential image, and the blood vessel path can be determined from the predicted differential image.

[0086] As an optional embodiment, step S302 involves sequentially inputting each differential image into a preset deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image, including the following steps:

[0087] Step S401: Determine the vanishing point of the blood vessel path corresponding to the differential image. Wherein, if the differential image is the edge of a blurred region, the vanishing point of the blood vessel path corresponding to the differential image is the vanishing point of the blood vessel path in the clear region. If the differential image is no longer the edge of a blurred region, the vanishing point of the blood vessel path corresponding to the differential image is the vanishing point of the blood vessel path in the predicted differential image obtained from the previous differential image.

[0088] Step S402: Input the vanishing point of the blood vessel path and the differential image into the target deep learning algorithm to obtain the predicted differential image after predicting the blood vessel path in the differential image.

[0089] For example, at the site of a cut, after image processing, we can find areas where the blood vessel direction is unclear. We can then perform blood vessel path analysis on these unclear points using differentiation. By differentiating along a certain point (the endpoint where the path disappears), we can obtain a differential image.

[0090] Then, the vanishing point of the blood vessel path and the differential image are input into the target deep learning algorithm to obtain the predicted differential image after predicting the blood vessel path in the differential image; thus, the purpose of predicting the blood vessel path in the differential image is achieved.

[0091] Step S303: Obtain the predicted finger vein sub-image based on all predicted differential images.

[0092] After obtaining all the predicted differential images, the predicted differential images can be stitched together to obtain the predicted finger vein sub-image.

[0093] As an optional embodiment, before sequentially inputting each differential image into a preset deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image, the method further includes the following steps:

[0094] Step S501: Obtain the training dataset and the validation dataset. The training dataset includes multiple sets of training data, and the validation dataset includes multiple sets of training data. Each set of training data includes corresponding vascular path vanishing points and training differential images.

[0095] Training data and validation datasets can be pre-collected or artificially constructed. The training dataset includes multiple sets of training data, and the validation dataset includes multiple sets of training data. The training data in the training dataset is used to train the deep learning algorithm to be trained, and the validation dataset includes multiple sets of training data to validate the trained deep learning algorithm.

[0096] Each set of training data includes corresponding vanishing points of blood vessel paths and training differential images. The vanishing point of the blood vessel path can be the starting position in the training differential image or the position in a clear image adjacent to the training differential image.

[0097] Step S502: Train the deep learning algorithm to be trained using the training dataset to obtain the trained deep learning algorithm.

[0098] After obtaining the training dataset, the deep learning algorithm to be trained can be trained using one or more sets of training data from the village-linked dataset to obtain the trained deep learning algorithm.

[0099] Step S503: After verifying the trained deep learning algorithm with the verification dataset and confirming that the trained deep learning algorithm meets the preset accuracy requirements, the trained deep learning algorithm is determined as the target deep learning algorithm.

[0100] After training the deep learning algorithm to be trained with a preset amount of training data and obtaining the trained deep learning algorithm, the trained deep learning algorithm can be verified using a validation dataset. If it is determined that the trained deep learning algorithm meets the preset accuracy requirements, the trained deep learning algorithm is determined as the target deep learning algorithm; otherwise, if it is determined that the trained deep learning algorithm does not meet the preset accuracy requirements, the trained deep learning algorithm is trained again with the training dataset until the final trained deep learning algorithm meets the preset accuracy requirements.

[0101] The preset accuracy requirement can be a threshold used to indicate the prediction accuracy of the deep learning algorithm after training, such as 95%, 90%, etc. It can be selected according to the actual application scenario, and will not be listed one by one here.

[0102] As an optional embodiment, step S104, where the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, determines the target finger as the specified finger, including the following steps:

[0103] Step S601: Determine the first vessel direction information of the historical finger vein sub-image and predict the second vessel direction information of the finger vein sub-image;

[0104] Step S602: If the first blood vessel direction information and the second blood vessel direction information meet a preset similarity threshold, the target finger is determined to be the specified finger.

[0105] After obtaining historical finger vein sub-images and predicted finger vein sub-images, the first vessel direction information of the historical finger vein sub-image and the second vessel direction information of the predicted finger vein sub-image can be determined.

[0106] Optionally, the first blood vessel orientation information can be obtained by determining the first feature points of the historical finger vein sub-image and then determining the orientation between each first feature point; similarly, the second blood vessel orientation information can be obtained by determining the second feature points of the predicted finger vein sub-image and then determining the orientation between each second feature point.

[0107] After obtaining the first and second blood vessel orientation information, similarity matching can be performed between the first and second blood vessel orientation information. If the first and second blood vessel orientation information meet a preset similarity threshold, the target finger can be identified as the specified finger.

[0108] The preset similarity can be a pre-set minimum value used to indicate the similarity between historical finger vein sub-images and predicted finger vein sub-images. For example, the preset similarity threshold can be 90%. In addition, other thresholds can be selected according to actual debugging, which will not be listed here.

[0109] As an optional embodiment, after determining that the target finger is the designated finger, the method further includes:

[0110] When the current finger vein image is used to unlock the lock, the motor of the lock is controlled to operate in order to unlock the lock.

[0111] After identifying the target finger as the designated finger, since the designated finger is a finger with pre-recorded historical finger vein images, it can be concluded that the information of the designated finger has been pre-recorded. Therefore, when the current finger vein image is used to unlock the lock, the motor of the lock can be controlled to unlock the lock.

[0112] like Figure 2 As shown, an application example applying any of the foregoing embodiments is provided:

[0113] 1. Acquire finger images through image acquisition and store them in the MCU. During recognition, read the current finger vein image of the user's finger and determine whether to proceed to the next image algorithm based on whether there are historical finger vein images in the MCU with 80% or more similar feature points.

[0114] 2. Compare the images with those stored in the MCU. Using an image segmentation algorithm, identify historical finger vein images with 80% (i.e., a preset matching threshold of 80%) similar feature points, and filter out the user IDs corresponding to these historical finger vein images. If a finger is injured, the wound may affect recognition during the process. Image comparison reveals that the current finger vein image shares at least 80% similarity in feature points with historical finger vein images. Further segmentation is then performed on the current finger vein image, separating identical regions and different regions. For example, if the blood vessels in the image are discontinuous or unclear, further segmentation is performed on the incomplete regions (i.e., blurred regions).

[0115] 3. For images with incomplete regions, differentiate along the path of the blood vessels at the edges of the same areas of the finger veins. Use deep learning algorithms to determine the direction of the blood vessels based on the direction of the local path. Compare and analyze the different directions of the blood vessels with the corresponding areas of historical finger vein images. If the similarity reaches a preset threshold, it can be further determined whether it is a recorded finger vein.

[0116] 4. Based on the threshold of blood vessel direction, if multiple incomplete regions can reach the preset similarity threshold with the previously screened ID-entry images (i.e., historical finger vein images) according to the segmentation algorithm, then this finger is considered to be the user's registered finger; and the motor can be controlled to unlock.

[0117] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0119] According to another aspect of the embodiments of this application, a finger vein-based finger recognition device is also provided for implementing the above-described finger recognition method based on finger veins. Figure 3 This is a structural block diagram of an optional finger recognition device based on finger veins according to an embodiment of this application, such as... Figure 3 As shown, the device may include:

[0120] Module 1 is used to acquire the current finger vein image obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with a wound.

[0121] Query module 2 is used to retrieve historical finger vein images that match the current finger vein image. The historical finger vein image is a finger vein image obtained by acquiring the image of the specified finger vein when there is no wound on the specified finger. The specified finger is a finger whose finger vein information has been pre-entered.

[0122] Prediction module 3 is used to predict the blood vessel direction in the blurred area of ​​the current finger vein image and obtain the predicted finger vein sub-image corresponding to the blurred area. The blurred area is the area in the current finger vein image where the blood vessel direction cannot be determined.

[0123] The determination module 4 is used to determine the target finger as the specified finger when the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold. The historical finger vein sub-image is the sub-image in the historical finger vein image that corresponds to the blurred area.

[0124] It should be noted that the acquisition module 1 in this embodiment can be used to perform the above step S101, the query module 2 in this embodiment can be used to perform the above step S102, the prediction module 3 in this embodiment can be used to perform the above step S103, and the determination module 4 in this embodiment can be used to perform the above step S104.

[0125] In addition to the modules described above, the apparatus in this embodiment may also include modules that perform any of the methods described in any of the foregoing embodiments of finger recognition methods based on finger veins.

[0126] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can operate in ways such as... Figure 1 The method shown can be implemented in either software or hardware within a hardware environment, where the hardware environment includes a network environment.

[0127] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described finger recognition method based on finger veins is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0128] According to another embodiment of this application, an electronic device is also provided, comprising: Figure 4 As shown, the electronic device may include: a processor 1501, a communication interface 1502, a memory 1503, and a communication bus 1504, wherein the processor 1501, the communication interface 1502, and the memory 1503 communicate with each other through the communication bus 1504.

[0129] Memory 1503 is used to store computer programs;

[0130] When processor 1501 executes the program stored in memory 1503, it performs the following steps:

[0131] Step S101: Obtain the current finger vein image obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with the current wound.

[0132] Step S102: Query and obtain historical finger vein images that match the current finger vein image. The historical finger vein image is a finger vein image obtained by acquiring the image of the specified finger vein when there is no wound on the specified finger. The specified finger is the finger whose finger vein information has been pre-entered.

[0133] Step S103: Predict the blood vessel direction in the blurred area of ​​the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred area, wherein the blurred area is the area in the current finger vein image where the blood vessel direction cannot be determined.

[0134] Step S104: If the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets the preset similarity threshold, the target finger is determined as the specified finger, wherein the historical finger vein sub-image is the sub-image in the historical finger vein image corresponding to the blurred region.

[0135] Optionally, in this embodiment, the communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used to represent it in the figure, but this does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned electronic device and other devices.

[0136] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0137] The processors mentioned above can be general-purpose processors, including but not limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; they can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0138] This application also provides a computer-readable storage medium, which includes a stored program, wherein the program executes the method steps of the above method embodiments when it runs.

[0139] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0140] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0141] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0142] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0143] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0144] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the solution provided in this embodiment, depending on actual needs.

[0145] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0146] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A finger recognition method based on finger veins, characterized in that, include: The current finger vein image is obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with a wound. The query retrieves historical finger vein images that match the current finger vein image. These historical finger vein images are obtained by capturing images of the specified finger veins when the specified finger has no wounds. The specified finger is a finger whose finger vein information has been pre-entered. The blood vessel direction in the blurred region of the current finger vein image is predicted to obtain a predicted finger vein sub-image corresponding to the blurred region, wherein the blurred region is the region in the current finger vein image where the blood vessel direction cannot be determined; If the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, the target finger is determined to be the designated finger, wherein the historical finger vein sub-image is the sub-image in the historical finger vein image corresponding to the blurred region.

2. The method according to claim 1, characterized in that, The query retrieves historical finger vein images that match the current finger vein image, including: Obtain all candidate finger vein images; Determine the matching degree between all feature points of each candidate finger vein image and all feature points of the current finger vein image; Based on the matching degree, the historical finger vein image is determined from all the candidate finger vein images, wherein the matching degree between the historical finger vein image and the current finger vein image is greater than or equal to a preset matching degree threshold.

3. The method according to claim 1, characterized in that, The step of predicting the blood vessel course in the blurred region of the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred region includes: Starting from the edge of the blurred region, the blurred region is differentiated according to the direction of the blood vessel path to obtain multiple consecutive differentiated images. After the multiple differentiated images are stitched together according to the direction of the blood vessel path, the blurred region can be restored. Each of the differential images is sequentially input into a preset target deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image; The predicted finger vein sub-image is obtained based on all the predicted differential images.

4. The method according to claim 3, characterized in that, The step of sequentially inputting each of the differential images into a preset deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image includes: Determine the vanishing point of the blood vessel path corresponding to the differential image, wherein, when the differential image is the edge of the blurred region, the vanishing point of the blood vessel path corresponding to the differential image is the vanishing point of the blood vessel path in the clear region; when the differential image is not the edge of the blurred region, the vanishing point of the blood vessel path corresponding to the differential image is the vanishing point of the blood vessel path in the predicted differential image obtained from the previous differential image. The vanishing point of the blood vessel path and the differential image are input into the target deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in the differential image.

5. The method according to claim 3, characterized in that, Before sequentially inputting each of the differential images into a preset deep learning algorithm to obtain a predicted differential image after predicting the blood vessel path in each differential image, the method further includes: Obtain a training dataset and a validation dataset, wherein the training dataset includes multiple sets of training data, and the validation dataset includes multiple sets of training data, each set of training data including corresponding blood vessel path vanishing points and training differential images; The deep learning algorithm to be trained is trained using the training dataset to obtain the trained deep learning algorithm. If the trained deep learning algorithm is validated using the validation dataset and it is determined that the trained deep learning algorithm meets the preset accuracy requirements, then the trained deep learning algorithm is identified as the target deep learning algorithm.

6. The method according to claim 1, characterized in that, The step of determining the target finger as the designated finger when the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold includes: Determine the first vessel orientation information of the historical finger vein sub-image and the second vessel orientation information of the predicted finger vein sub-image; If the preset similarity threshold is met between the first blood vessel direction information and the second blood vessel direction information, the target finger is determined to be the designated finger.

7. The method according to any one of claims 1 to 6, characterized in that, After determining that the target finger is the designated finger, the method further includes: When the current finger vein image is used to unlock the lock, the motor of the lock is controlled to operate in order to unlock the lock.

8. A finger recognition device based on finger veins, characterized in that, include: The acquisition module is used to acquire the current finger vein image obtained by image acquisition of the target finger vein of the target object, wherein the target finger vein is the finger vein of the target finger with a wound. The query module is used to query and obtain historical finger vein images that match the current finger vein image. The historical finger vein image is a finger vein image obtained by acquiring the image of the specified finger vein when the specified finger has no wound. The specified finger is a finger whose finger vein information has been pre-entered. The prediction module is used to predict the blood vessel direction in the blurred area of ​​the current finger vein image to obtain a predicted finger vein sub-image corresponding to the blurred area, wherein the blurred area is the area in the current finger vein image where the blood vessel direction cannot be determined; The determination module is used to determine the target finger as the specified finger when the similarity between the historical finger vein sub-image and the predicted finger vein sub-image meets a preset similarity threshold, wherein the historical finger vein sub-image is a sub-image in the historical finger vein image corresponding to the blurred region.

9. An electronic device comprising a processor, a communication interface, a memory, and a communication bus, wherein, The processor, the communication interface, and the memory communicate with each other via the communication bus, characterized in that... The memory is used to store computer programs; The processor is configured to perform the method steps of any one of claims 1 to 7 by running the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the steps of the method described in any one of claims 1 to 7 when it is run.

Images