Feature point recognition method and electrical apparatus

Abstract

The application provides a feature point recognition method and an electric device. The feature point recognition method comprises the following steps: acquiring feature points of images, wherein the images at least comprise a first image and a second image; determining a second image feature point region according to the feature points of the first image; and determining effective feature points of the first image according to the second image feature point region and the feature points of the first image. The feature point recognition method provided by the application improves the flexibility of feature point extraction, reduces the number of feature points in the recognition process, and improves the recognition efficiency by screening the feature points of the images to obtain the effective feature points.

Description

Technical Field

[0001] This application relates to the field of palm vein recognition technology, and in particular to a feature point recognition method and electrical equipment. Background Technology

[0002] In related technologies, existing palm vein recognition methods typically divide the image into at least two sub-regions and then extract feature points from each sub-region. This feature point extraction method results in a large number of feature points being extracted, and the sub-regions are relatively fixed, increasing computational complexity and reducing the flexibility of feature point extraction. Summary of the Invention

[0003] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a feature point recognition method and an electrical device. This feature point recognition method, by filtering feature points in an image to obtain effective feature points, improves the flexibility of feature point extraction, reduces the number of feature points required in the recognition process, and improves recognition efficiency.

[0004] To achieve the above objectives, a feature point recognition method is proposed according to a first aspect of this application, comprising: acquiring feature points of an image, wherein the image includes at least a first image and a second image; determining a feature point region of a second image based on the feature points of the first image; and determining valid feature points of the first image based on the feature point region of the second image and the feature points of the first image.

[0005] According to the first aspect of this application, a feature point recognition method is proposed. By filtering the feature points of an image to obtain effective feature points, the flexibility of feature point extraction is improved, the number of feature points in the recognition process is reduced, and the recognition efficiency is improved.

[0006] In some examples of this application, determining the feature point region of the second image based on the feature points of the first image includes: mapping the feature points of the first image onto the second image to obtain mapping points; and determining the feature point region of the second image based on the mapping points.

[0007] In some examples of this application, determining the second image feature point region based on the mapping point includes: determining a square region centered on the mapping point and with a certain side length as the second image feature point region.

[0008] In some examples of this application, determining the effective feature points of the first image based on the feature point region of the second image and the feature points of the first image includes: determining matching points in the feature point region of the second image based on the feature point region of the second image; determining the L1 norm between the matching point and the feature points of the first image based on the matching point; determining a first L1 norm and a second L1 norm based on the L1 norm between the matching point and the feature points of the first image; and determining the effective feature points of the first image based on the ratio of the first L1 norm to the second L1 norm.

[0009] In some examples of this application, the method further includes: determining the feature point score of the first image based on the valid feature points of the first image; and determining the feature point ranking of the first image based on the feature point score of the first image.

[0010] In some examples of this application, the method further includes: obtaining the similarity between the first image and the second image; and filtering the first image based on the similarity between the first image and the second image.

[0011] In some examples of this application, the method further includes: determining a first matching point of the second image based on the first L1 norm; determining the Hamming distance based on the valid feature points of the first image and the first matching point of the second image; and determining the unlocked state based on the Hamming distance and a first numerical value.

[0012] According to a second aspect of this application, an electrical device is provided, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, it implements the steps of the feature point recognition method provided in this embodiment.

[0013] According to a third aspect of this application, a computer-readable storage medium is provided, on which a computer program or instructions are stored, which, when executed by a processor, implement the steps of the feature point recognition method provided in this embodiment.

[0014] According to a fourth aspect of this application, a computer program product is also provided, including a computer program or instructions, which, when executed by a processor, implement the steps of the feature point recognition method provided in this embodiment.

[0015] According to a fifth aspect of this application, a robot is also provided, including the electrical equipment provided in the second aspect embodiment of this application, or the computer-readable storage medium provided in the third aspect embodiment of this application, or the computer program product provided in the fourth aspect embodiment of this application.

[0016] According to a sixth aspect of this application, a vehicle is also provided, including the electrical equipment provided in the second aspect embodiment of this application, or the robot provided in the fifth aspect embodiment of this application.

[0017] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0020] Figure 1 This is a flowchart illustrating a feature point recognition method provided in some embodiments of this application;

[0021] Figure 2 This is a flowchart illustrating a feature point recognition method provided in some embodiments of this application;

[0022] Figure 3 This is a flowchart illustrating a feature point recognition method provided in some embodiments of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0024] The following is for reference. Figure 1-3 A feature point recognition method according to embodiments of this application is described in detail.

[0025] In some embodiments, such as Figure 1 As shown, the feature point recognition method includes the following steps:

[0026] S101, Obtain feature points of the image, the image includes at least the first image and the second image.

[0027] In some embodiments, feature points of an image need to be acquired before performing the feature point recognition method. The image includes at least a first image and a second image. The first image and the second image can refer to a single image or a class of images; no further limitation is made here. For example, when the first image and the second image refer to a class of images, the first image can refer to the image to be tested, and the second image can refer to the comparison image compared with the image to be tested.

[0028] In some specific embodiments, the feature point recognition method is mainly used for palm vein image recognition; therefore, the first image and the second image mainly refer to the acquired palm vein images. In other embodiments, the feature point recognition method of this application can also be applied to scenarios involving feature point recognition, such as facial recognition and fingerprint recognition.

[0029] S102, determine the feature point region of the second image based on the feature points of the first image.

[0030] In some embodiments, feature points of a first image are obtained, and feature point regions of a second image can be determined based on the feature points of the first image.

[0031] S103, determine the effective feature points of the first image based on the feature point region of the second image and the feature points of the first image.

[0032] In some embodiments, the effective feature points of the first image can be determined based on the feature point region of the second image and the feature points of the first image.

[0033] In summary, the feature point recognition method proposed in this application improves the flexibility of feature point extraction, reduces the number of feature points in the recognition process, and improves recognition efficiency by filtering the feature points of the image to obtain effective feature points.

[0034] In some embodiments, such as Figure 2 As shown, step S101, determining the feature point region of the second image based on the feature points of the first image, includes the following steps:

[0035] S201, map the feature points of the first image onto the second image to obtain the mapped points.

[0036] In some embodiments, after obtaining the feature points of the first image and the second image, it is necessary to map the feature points of the first image onto the second image to obtain the mapped points of the feature points of the first image on the second image. It should be noted that the mapped points of the feature points of the first image on the second image refer to the alignment points of the feature points of the first image on the second image. Alignment points are determined by comparing the feature points on the first image and the feature points on the second image. If the X-axis distance and Y-axis distance between the coordinates of the feature points on the second image and the coordinates of the feature points on the first image are both within a certain threshold, then the feature points on the second image are considered to be the alignment points of the feature points on the first image.

[0037] S202, Determine the second image feature point region based on the mapping points.

[0038] In some embodiments, after confirming the mapping points, the second image feature point region can be determined.

[0039] In some embodiments, a square region centered on the mapping point and with a certain side length is defined as the second image feature point region. This certain value can be a length value, a width value, or a pixel value, etc., and is not limited here. That is, the second image feature point region can be a range within a 32*32 area centered on the mapping point, where the unit of the value 32 is pixels.

[0040] In other embodiments, the shape of the second image feature point region is not limited. The preferred embodiment is a square region with a certain side length centered on the mapping point. It can also be a rectangular region centered on the mapping point or a circular region centered on the mapping point. No further limitations are imposed here.

[0041] In some embodiments, such as Figure 3 As shown, step S103, which determines the effective feature points of the first image based on the feature point region of the second image and the feature points of the first image, further includes the following steps:

[0042] S301, Based on the feature point region of the second image, determine the matching point in the feature point region of the second image.

[0043] In some embodiments, after obtaining the second image feature point region, the second image feature points existing in the second image feature point region are determined. The second image feature points located in the second image feature point region are the matching points in the second image feature point region. It can be considered that the feature points in the second image feature point region are the matching points that match the feature points of the first image.

[0044] S302, Based on the matching point, determine the L1 norm between the matching point and the feature points of the first image.

[0045] In some embodiments, after determining the matching point in the feature point region of the second image, it is necessary to determine the L1 norm between the matching point and the feature points of the first image. It should be noted that the L1 norm, also known as the Manhattan distance or the sum of absolute values, is the sum of the absolute values ​​of all elements in a vector.

[0046] S303, determine the first L1 norm and the second L1 norm based on the L1 norm between the matching point and the feature points of the first image.

[0047] In some embodiments, generally, since there are multiple matching points, the L1 norm between the numerous matching points and the feature points of the first image can be used to determine the first L1 norm and the second L1 norm. It should be noted that the first L2 norm is the minimum value among all L1 norms, and the second L1 norm is the second smallest value among all L1 norms.

[0048] S304, determine the effective feature points of the first image based on the ratio of the first L1 norm to the second L1 norm.

[0049] In some embodiments, after obtaining the values ​​of the first L1 norm and the second L1 norm, the ratio of the first L1 norm to the second L1 norm can be determined, and the effective feature points of the first image can be determined based on the ratio of the first L1 norm to the second L1 norm.

[0050] In some embodiments, a first threshold is also required. When the ratio of the first L1 norm to the second L1 norm is less than the first threshold, valid feature points of the first image are determined. The first threshold is an indefinite value less than or equal to 1. A threshold closer to 1 indicates a more stringent selection process for valid feature points in the first image, while a threshold further away from 1 indicates a more lenient selection process. Generally, the first threshold can be set to 0.8 or 0.9.

[0051] Therefore, when a feature point in the first image simultaneously satisfies the following conditions: a corresponding matching point can be found in the feature point region of the second image, and the ratio of the first L1 norm to the second L1 norm between the matching point and the feature point in the first image is less than a first threshold, the feature point in the first image is considered a valid feature point. It should be noted that the coordinate systems of the first and second images are the same.

[0052] In some embodiments, the feature point recognition method of this application further includes the following steps:

[0053] Based on the effective feature points of the first image, determine the feature point score of the first image;

[0054] The feature point ranking of the first image is determined based on the feature point scores of the first image.

[0055] In some embodiments, feature point scores of the first image are determined based on valid feature points. Feature points in the first image are divided into valid feature points and invalid feature points, where a valid feature point is scored as one point and an invalid feature point is scored as zero points. Thus, the feature point scores of the first image can be obtained. Based on the feature point scores of the first image, the feature points of the first image are sorted, and invalid feature points with scores of zero are filtered out.

[0056] In some embodiments, when the second image consists of more than one image, the above method steps are performed on the images in the first image and the images in the second image in turn to score them. This results in feature points in the first image being valid feature points in some images of the second image and invalid feature points in others, leading to a hierarchical scoring system. For example, if the first image is one image (hereinafter referred to as image A) and the second image is four images (hereinafter referred to as images B, C, D, and E), a feature point on image A is compared with a feature point on image B. Through the above steps, it is determined that the feature point is valid for image B. However, after the same steps, the same feature point on image A is not a valid feature point in images C, D, or E; therefore, this feature point is scored as 1 point. Similarly, there are feature points scored as 4, 3, and 2 points. These feature points are sorted and stored according to their scores, and we only need to filter out feature points with a score of 0.

[0057] In some embodiments, the feature point recognition method of this application further includes the following steps:

[0058] Obtain the similarity between the first image and the second image; filter the first image based on the similarity between the first image and the second image.

[0059] In some embodiments, a first-stage rapid screening can be performed on the first image and the second image. The similarity between the first image and the second image is calculated. Only when the similarity is extremely high will a second-stage feature point comparison be performed, thereby filtering out images with low similarity and improving recognition efficiency.

[0060] In some embodiments, the feature point recognition method of this application further includes the following steps:

[0061] Based on the first L1 norm, determine the first matching point of the second image;

[0062] The Hamming distance is determined based on the effective feature points of the first image and the first matching point of the second image;

[0063] The unlock status is determined based on the Hamming distance and the first value.

[0064] In some embodiments, the matching point of the second image corresponding to the first L1 norm is the first matching point of the second image. Based on the valid feature points of the first image and the feature point descriptors of the pixels of the first matching point in the second image, the Hamming distance can be determined. Based on the Hamming distance and the first numerical value, the unlocked state can be determined. The Hamming distance is a measure of the difference between two strings, defined as the number of different characters at the same position. It should be noted that there are various ways to calculate the feature point descriptor, which will not be elaborated here. This application uses LBP (Local Binary Pattern).

[0065] Furthermore, if there is only one pair of valid feature points in the first image and the first matching point in the second image, and the Hamming distance is less than a first value, then unlocking is considered successful. Conversely, if the Hamming distance is greater than the first value, unlocking fails. When the Hamming distance equals the first value, either unlocking is successful or unsuccessful, depending on the user's settings. If there are at least two pairs of valid feature points in the first image and the first matching point in the second image, and at least one pair satisfies the condition that the Hamming distance is less than the first value, then unlocking is considered successful. If none of the pair satisfies the condition that the Hamming distance is less than the first value, then unlocking fails.

[0066] In some specific embodiments, the concept of segmentation is adopted. The first image is the verification image, and the second image is the registration image. Each feature point on the verification image is mapped to the registration image. Matching calculation is performed within a 32*32 area centered on the point. Then, the ratio of the optimal distance to the second-best distance of each matching pair is calculated. If it is less than a threshold of 0.8, it is stored as a feature pair to be sorted; otherwise, it is discarded. All stored points in the database are iterated. For each stored feature pair to be sorted, feature point alignment is performed first. Then, the Hamming distance of each pair is downsampled and calculated and sorted. The first-ranked matching pair is used to calculate the Hamming distance of the original 96*96 feature map and compared with the final threshold to output the unlocking result.

[0067] In some specific embodiments, the solution of this application can be applied to robots, specifically to palm vein recognition. Before starting or interacting, the robot acquires images of the user's palm using a fixed RGB camera or near-infrared camera module. An image sharpness evaluation function (such as the Brenner gradient) is used to automatically remove blurry or occluded images, ensuring the quality of the recognition image. The Brenner gradient function is used to calculate the square of the gray-level difference between two adjacent pixels, and the summation over the horizontal and vertical textures is used as the final image quality evaluation value, as shown in the following formula:

[0068]

[0069] Where x and y represent the coordinates of the pixel, and f(x,y) represents the grayscale value of the image at coordinates x and y, respectively; the first part of the formula calculates the horizontal texture, and the second part calculates the vertical texture.

[0070] This application also provides an electrical device, including a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the steps of the feature point recognition method provided in this embodiment.

[0071] It should be noted that the above-mentioned electrical equipment can be any conventional equipment that requires electricity, such as, but not limited to, controllers, vehicles, skateboard chassis, ships, drones, robots, mobile phones, computers, air conditioners, refrigerators, washing machines, microwave ovens, printers, fax machines, etc.

[0072] This application also provides a computer-readable storage medium storing a computer program or instructions that, when executed by a processor, cause the processor to be configured to perform the steps of the feature point recognition method provided in this embodiment.

[0073] This application also provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps of the feature point recognition method provided in this application.

[0074] This application also provides a robot, including the electrical equipment provided in this application embodiment.

[0075] This application also provides a vehicle, including the electrical equipment provided in this application embodiment, or the robot provided in this application embodiment. In this embodiment, the vehicle may be a gasoline-powered vehicle, a plug-in hybrid electric vehicle, or a new energy vehicle, etc., and this application does not specifically limit it.

[0076] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0078] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0079] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0080] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0081] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0082] Computer-readable media include both permanent and non-permanent, removable and non-removable media, which can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient media, such as modulated communication signals and carrier waves.

[0083] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0084] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

[0085] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0086] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0087] The embodiments, implementation methods, and related technical features of this application can be combined and substituted for each other without conflict.

[0088] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application without departing from the scope of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A feature point recognition method, characterized in that, include: Acquire feature points of an image, wherein the image includes at least a first image and a second image; Based on the feature points of the first image, determine the feature point region of the second image; The effective feature points of the first image are determined based on the feature point region of the second image and the feature points of the first image.

2. The method according to claim 1, characterized in that, Determining the feature point region of the second image based on the feature points of the first image includes: The feature points of the first image are mapped onto the second image to obtain the mapped points; The second image feature point region is determined based on the mapping points.

3. The method according to claim 2, characterized in that, Determining the second image feature point region based on the mapping points includes: A square region centered at the mapping point and with a certain side length is determined as the second image feature point region.

4. The method according to claim 1, characterized in that, The step of determining the effective feature points of the first image based on the feature point region of the second image and the feature points of the first image includes: Based on the second image feature point region, determine the matching point in the second image feature point region; Based on the matching point, determine the L1 norm between the matching point and the feature point of the first image; The first L1 norm and the second L1 norm are determined based on the L1 norm between the matching point and the feature point of the first image; The effective feature points of the first image are determined based on the ratio of the first L1 norm to the second L1 norm.

5. The method according to claim 4, characterized in that, Also includes: Based on the valid feature points of the first image, determine the feature point score of the first image; The feature point ranking of the first image is determined based on the feature point scores of the first image.

6. The method according to claim 1, characterized in that, Also includes: Obtain the similarity between the first image and the second image; The first image is filtered based on the similarity between the first image and the second image.

7. The method according to claim 4, characterized in that, Also includes: Based on the first L1 norm, determine the first matching point of the second image; The Hamming distance is determined based on the effective feature points of the first image and the first matching point of the second image; The unlock status is determined based on the Hamming distance and the first value.

8. An electrical appliance, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7.

Images