Face recognition method and system, computer equipment and readable storage medium

Abstract

The invention relates to a face recognition method and system, computer equipment and a readable storage medium. The face recognition method comprises the following steps: acquiring face image data; repairing the face image data to remove interference data; performing feature extraction on the repaired face data to obtain a feature vector set; segmenting the feature vector set by using a cloud platform to obtain a plurality of sub-feature vector sets; respectively distributing each sub-feature vector set to each processing node of the cloud platform to carry out face recognition parallel processing so as to obtain a plurality of parallel processing results; and combining the plurality of parallel processing results to obtain a face recognition result. According to the method, the interference data in the face image data are removed by repairing the face before the face is recognized, so that the method has relatively high anti-interference performance, and the accuracy of a recognition result is ensured; in addition, through parallel processing, the processing speed of face recognition can be greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of face recognition technology, and in particular to a face recognition method, system, computer equipment and readable storage medium. Background Art

[0002] In the current technical field, face recognition technology has been widely used in many fields such as security monitoring, identity authentication, and passenger flow analysis. However, with the diversification of application scenarios, traditional face recognition systems face problems such as slow processing speed, low accuracy, and poor anti-interference ability. Especially in large-scale image processing, existing technologies often rely on local computing resources, which limits their processing capabilities and efficiency. Summary of the invention

[0003] Based on this, it is necessary to provide a face recognition method, system, computer device and readable storage medium to address the above technical problems.

[0004] In a first aspect, the present invention provides a face recognition method, the face recognition method comprising:

[0005] Acquire facial image data, wherein the facial image data includes interference data, and the interference data includes one of light interference and occlusion interference;

[0006] Repairing the facial image data to remove the interference data;

[0007] Perform feature extraction on the restored face data to obtain a feature vector set;

[0008] Using a cloud platform to segment the feature vector set to obtain a plurality of sub-feature vector sets;

[0009] Distributing each of the sub-feature vector sets to each processing node of the cloud platform for parallel face recognition processing to obtain multiple parallel processing results;

[0010] The multiple parallel processing results are combined to obtain a face recognition result.

[0011] Optionally, repairing the facial image data to remove the interference data includes:

[0012] Identifying interference types of interference data in the face image data;

[0013] Based on the interference type of the interference data, the facial image data is repaired to remove the interference data.

[0014] Optionally, identifying whether the facial image data includes light interference includes:

[0015] Obtaining a histogram of the face image data;

[0016] Obtaining uniformity of the histogram;

[0017] It is determined whether the uniformity of the histogram is greater than a preset threshold, and if so, it is determined that light interference exists in the face image data.

[0018] Optionally, if the facial image data includes light interference, repairing the facial image data to remove the interference data includes: compensating and repairing the facial image data based on the following formula:

[0019]

[0020] Where, I'(x, y) is the pixel value of the compensated face image data at the coordinate (x, y); I(x, y) is the pixel value of the original face image data at the coordinate (x, y); λ is an adjustment parameter used to control the compensation intensity; H is the uniformity of the histogram;

[0021] L is the number of gray levels of the face image data; h(i) is the frequency of the i-th gray level; N is the total number of pixels in the face image data.

[0022] Optionally, identifying whether the face image data includes occlusion interference includes:

[0023] Obtaining key feature points in the face image data, and determining whether there is a non-skin area outside the eye area;

[0024] If the key feature point is missing or there is a non-skin area outside the eye area, it is determined that there is occlusion interference in the face image data.

[0025] Optionally, if the face image data includes occlusion interference, repairing the face image data to remove the interference data includes:

[0026] Identifying and marking the occluded area in the facial image data;

[0027] Image restoration technology is used to repair the occluded areas identified and marked.

[0028] Optionally, the cloud platform divides the feature vector set based on a distributed computing framework to obtain multiple sub-feature vector sets; based on the distributed computing framework, each of the sub-feature vector sets is respectively distributed to each processing node of the cloud platform for parallel face recognition processing, and each processing node compares the feature vector in each sub-feature vector set with the target feature vector to obtain multiple parallel processing results.

[0029] In a second aspect, the present invention further provides a face recognition system, the face recognition system comprising:

[0030] A face image data acquisition module, used to acquire face image data, wherein the face image data includes interference data, and the interference data includes one of light interference and occlusion interference;

[0031] A repair module, repairing the face image data to remove the interference data;

[0032] A feature extraction module is used to extract features from the restored face data to obtain a feature vector set;

[0033] The cloud platform is used to segment the feature vector set to obtain a plurality of sub-feature vector sets; and distribute each of the sub-feature vector sets to each processing node of the cloud platform for parallel face recognition processing to obtain a plurality of parallel processing results;

[0034] The merging processing module is used to merge the multiple parallel processing results to obtain a face recognition result.

[0035] In a third aspect, the present invention further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the face recognition method described in the first aspect when executing the computer program.

[0036] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the face recognition method as described in the first aspect are implemented.

[0037] The above-mentioned face recognition method, system, computer device and readable storage medium of the present invention have strong anti-interference performance and ensure the accuracy of recognition results by repairing before face recognition and removing interference data in face image data; in addition, through parallel processing, the processing speed of face recognition can be greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flowchart of a face recognition method is provided in one embodiment of the present invention.

[0039] Figure 2 A structural block diagram of a face recognition system provided in another embodiment of the present invention.

[0040] Figure 3 This is a diagram of the internal structure of a computer device provided in yet another embodiment of the present invention.

[0041] Explanation of the accompanying drawings: 10. Face image data acquisition module; 11. Restoration module; 12. Feature extraction module; 13. Cloud platform; 14. Merging processing module. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0043] In one embodiment, see Figure 1 The present invention provides a face recognition method, which may include the following steps: S10~S50.

[0044] S10: Acquire facial image data, wherein the facial image data includes interference data, and the interference data includes one of light interference and occlusion interference.

[0045] S20: Repairing the facial image data to remove the interference data.

[0046] S30: Extract features from the restored face data to obtain a feature vector set.

[0047] S40: Using a cloud platform to segment the feature vector set to obtain a plurality of sub-feature vector sets.

[0048] S50: Allocate each of the sub-feature vector sets to each processing node of the cloud platform for parallel face recognition processing to obtain multiple parallel processing results.

[0049] S60: merging the plurality of parallel processing results to obtain a face recognition result.

[0050] By repairing the face before recognition and removing the interference data in the face image data, it has strong anti-interference performance and ensures the accuracy of the recognition results; in addition, through parallel processing, the processing speed of face recognition can be greatly improved.

[0051] As an example, in step S10, facial image data is obtained, and the facial image data includes interference data, and the interference data includes one of light interference and occlusion interference.

[0052] As an example, a camera or other acquisition device may be used to acquire facial image data.

[0053] As an example, when acquiring facial image data, due to the influence of the environment (for example, whether the collection scene is indoors or outdoors), there may be light interference when acquiring the facial image data. For example, the light may be too dark when acquiring facial image data indoors, and the light may be too strong when acquiring facial image data outdoors. This will cause light interference in the collected facial image data.

[0054] In addition, during the acquisition process, due to different scenes, there may be some objects that partially block the face, resulting in occlusion interference in the collected face image data.

[0055] It should be noted that both light interference and occlusion interference will affect the quality of the acquired facial image data, thereby affecting subsequent face recognition, resulting in inaccurate or failed subsequent face recognition.

[0056] In step S20, refer to Figure 1 In step S20, the face image data is repaired to remove the interference data.

[0057] As an example, step S20 may include the following steps: S201 to S202.

[0058] S201: Identify the interference type of the interference data in the face image data.

[0059] S202: Repairing the facial image data based on the interference type of the interference data to remove the interference data.

[0060] As an example, before identifying the interference type of the interference data in the face image data, the step of preprocessing the face image data may also be included, and step S20 is to identify the interference type of the interference data in the preprocessed face image data. The preprocessing may include but is not limited to grayscale conversion and filtering denoising.

[0061] As an example, in step S201, identifying whether the facial image data includes light interference may include the following steps: obtaining a histogram of the facial image data; obtaining the uniformity of the histogram; determining whether the uniformity of the histogram is greater than a preset threshold, and if so, determining that light interference exists in the facial image data.

[0062] Specifically, the uniformity of the histogram can be obtained based on the following formula:

[0063] L is the number of gray levels of the face image data; h(i) is the frequency of the i-th gray level; N is the total number of pixels in the face image data.

[0064] As an example, in step S201, identifying whether the facial image data includes occlusion interference may include: obtaining key feature points in the facial image data, and determining whether there is a non-skin area outside the eye area; if the key feature points are missing or there is a non-skin area outside the eye area, it is determined that there is occlusion interference in the facial image data.

[0065] Specifically, the key feature points may be obtained by extracting the facial image data; and whether there is a non-skin area outside the eye area may be determined by comparing the color and / or grayscale in the facial image data.

[0066] As an example, the preset threshold can be set according to actual needs, and is no longer similar here.

[0067] As an example, if the face image data includes light interference, in step S202, repairing the face image data to remove the interference data may include: compensating and repairing the face image data based on the following formula:

[0068]

[0069] Wherein, I'(x, y) is the pixel value of the compensated face image data at the coordinate (x, y); I(x, y) is the pixel value of the original face image data at the coordinate (x, y); λ is an adjustment parameter used to control the compensation intensity; H is the uniformity of the histogram.

[0070] As an example, if the facial image data includes occlusion interference, in step S202, the repairing of the facial image data to remove the interference data may include the following steps: identifying and marking the occluded area in the facial image data; and repairing the identified and marked occluded area using image repair technology.

[0071] Specifically, any existing image annotation technology may be used to identify and annotate the occluded area in the face image data. Any existing image restoration technology may be used to restore the identified and annotated occluded area.

[0072] In step S30, refer to Figure 1 In step S30, feature extraction is performed on the restored face data to obtain a feature vector set.

[0073] As an example, in step S30, feature extraction can be performed on the restored face data based on a deep learning algorithm, etc., to obtain a feature vector set. The specific method of extracting features to obtain a feature vector set is known to those skilled in the art and will not be described here.

[0074] In step S40, refer to Figure 1 In step S40, the feature vector set is segmented using a cloud platform to obtain multiple sub-feature vector sets.

[0075] As an example, the cloud platform can segment the feature vector set based on a distributed computing framework (such as Apache Spark or Hadoop) to obtain multiple sub-feature vector sets. The distributed computing framework can segment the large-scale data of the feature vector set into multiple small blocks (i.e., sub-feature vector sets).

[0076] In step S50, refer to Figure 1 In step S50, each of the sub-feature vector sets is respectively distributed to each processing node of the cloud platform for parallel processing of face recognition to obtain multiple parallel processing results.

[0077] As an example, based on the distributed computing framework, each of the sub-feature vector sets can be respectively distributed to each processing node of the cloud platform for parallel processing of face recognition. Each processing node compares the feature vector in each of the sub-feature vector sets with the target feature vector to obtain multiple parallel processing results.

[0078] Specifically, each processing node processes the comparison and identification of different sub-feature vector sets respectively. By means of parallel processing, the data processing speed and system throughput can be significantly improved, and the processing speed of face recognition can be improved.

[0079] As an example, each of the processing nodes may use cosine similarity or Euclidean distance to compare the feature vectors in each of the sub-feature vector sets with the target feature vector to obtain a plurality of parallel processing results.

[0080] In step S60, refer to Figure 1 In step S60, the plurality of parallel processing results are combined to obtain a face recognition result.

[0081] As an example, the existing image fusion and merging technology may be used to merge the multiple parallel processing results to obtain the final face recognition result.

[0082] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0083] In another embodiment, see Figure 2 The present invention also provides a face recognition system, which may include: a face image data acquisition module 10, a repair module 11, a feature extraction module 12, a cloud platform 13 and a merging processing module 14; wherein the face image data acquisition module 10 is used to acquire face image data, the face image data includes interference data, and the interference data includes one of light interference and occlusion interference; the repair module 11 repairs the face image data to remove the interference data; the feature extraction module 12 is used to extract features from the repaired face data to obtain a feature vector set; the cloud platform 13 is used to segment the feature vector set to obtain multiple sub-feature vector sets; and each of the sub-feature vector sets is respectively allocated to each processing node of the cloud platform for parallel face recognition processing to obtain multiple parallel processing results; the merging processing module 14 is used to merge multiple parallel processing results to obtain a face recognition result.

[0084] For the specific definition of the face recognition system, please refer to the definition of the face recognition method above, which will not be repeated here. Each module in the above face recognition system can be implemented in whole or in part by software, hardware and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0085] In another embodiment, the present invention further provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store historical work data, real-time work data and other data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it is implemented as follows Figure 1 And the face recognition method provided in the corresponding embodiment.

[0086] In another embodiment, the present invention further provides a computer device, which may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it is implemented as follows Figure 1 The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0087] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0088] In another embodiment of the present invention, the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following is implemented: Figure 1 And the face recognition method provided in the corresponding embodiment.

[0089] In yet another embodiment, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following is implemented: Figure 1 And the face recognition method provided in the corresponding embodiment.

[0090] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0091] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0092] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A face recognition method, characterized in that: include: Acquire facial image data, wherein the facial image data includes interference data, and the interference data includes one of light interference and occlusion interference; Repairing the facial image data to remove the interference data; Perform feature extraction on the restored face data to obtain a feature vector set; Using a cloud platform to segment the feature vector set to obtain a plurality of sub-feature vector sets; Distributing each of the sub-feature vector sets to each processing node of the cloud platform for parallel face recognition processing to obtain multiple parallel processing results; The multiple parallel processing results are combined to obtain a face recognition result.

2. The face recognition method according to claim 1, characterized in that: Repairing the facial image data to remove the interference data includes: Identifying interference types of interference data in the face image data; Based on the interference type of the interference data, the facial image data is repaired to remove the interference data.

3. The face recognition method according to claim 1, characterized in that: Identifying whether the face image data includes light interference includes: Obtaining a histogram of the face image data; Obtaining uniformity of the histogram; It is determined whether the uniformity of the histogram is greater than a preset threshold, and if so, it is determined that light interference exists in the face image data.

4. The face recognition method according to claim 3, characterized in that: If the face image data includes light interference, repairing the face image data to remove the interference data includes: compensating and repairing the face image data based on the following formula: Where, I'(x, y) is the pixel value of the compensated face image data at the coordinate (x, y); I(x, y) is the pixel value of the original face image data at the coordinate (x, y); λ is an adjustment parameter used to control the compensation intensity; H is the uniformity of the histogram; L is the number of gray levels of the face image data; h(i) is the frequency of the i-th gray level; N is the total number of pixels in the face image data.

5. The face recognition method according to claim 1, characterized in that: Identifying whether the face image data includes occlusion interference includes: Obtaining key feature points in the face image data, and determining whether there is a non-skin area outside the eye area; If the key feature point is missing or there is a non-skin area outside the eye area, it is determined that there is occlusion interference in the face image data.

6. The face recognition method according to claim 5, characterized in that: If the face image data includes occlusion interference, repairing the face image data to remove the interference data includes: Identifying and marking the occluded area in the facial image data; Image restoration technology is used to repair the occluded areas identified and marked.

7. The face recognition method according to claim 1, characterized in that: The cloud platform divides the feature vector set based on a distributed computing framework to obtain multiple sub-feature vector sets; based on the distributed computing framework, each of the sub-feature vector sets is respectively distributed to each processing node of the cloud platform for parallel processing of face recognition, and each processing node compares the feature vector in each sub-feature vector set with the target feature vector to obtain multiple parallel processing results.

8. A face recognition system, characterized in that: The face recognition system comprises: A face image data acquisition module, used to acquire face image data, wherein the face image data includes interference data, and the interference data includes one of light interference and occlusion interference; A repair module, repairing the face image data to remove the interference data; A feature extraction module is used to extract features from the restored face data to obtain a feature vector set; The cloud platform is used to segment the feature vector set to obtain a plurality of sub-feature vector sets; and distribute each of the sub-feature vector sets to each processing node of the cloud platform for parallel face recognition processing to obtain a plurality of parallel processing results; The merging processing module is used to merge the multiple parallel processing results to obtain a face recognition result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images