A smart campus identity management system and method based on face recognition

Abstract

The present invention belongs to the technical field of face recognition. The present invention discloses a smart campus identity management system and method based on face recognition, including: collecting a sequence of face images and corresponding identity information, performing dynamic feature tracking on the sequence of face images to obtain a sequence of feature marker maps; performing skin state analysis on the sequence of feature marker maps to obtain a skin state feature set; performing expression analysis on the skin state feature set to obtain static features and micro-expression features, and performing micro-deformation compensation on the micro-expression features to obtain micro-motion compensation features; constructing an identity recognition model based on the compensation features, static features and identity information, and realizing accurate identification of personnel based on the identity recognition model; greatly improving the accuracy and efficiency of identity recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of face recognition, and more specifically, to an intelligent campus identity management system and method based on face recognition. Background Art

[0002] With the rapid development of artificial intelligence and biometric technologies, face recognition has been widely applied in fields such as security verification, identity recognition, and behavior analysis. In practical applications, traditional face recognition systems still face a series of challenges. For example, in dynamic scenarios such as speaking and laughing, due to continuous deformation of facial muscles, feature points are prone to drift. In this case, it is difficult for traditional face recognition systems to accurately extract facial features, resulting in incorrect feature point matching, thus affecting the accuracy of identity recognition. In addition, for special situations such as local scars and temporary skin lesions on the face, the adaptability of existing systems is poor. Facial abnormal deformations (such as pimples, scars, or other short-term skin problems) often significantly reduce the accuracy of feature point extraction and matching, which may lead to misrecognition or rejection in a large-scale campus environment. Most existing face recognition algorithms are based on feature extraction from static images and rarely consider the influence of dynamic factors such as expressions and postures. However, in practical applications, facial posture changes, expressions such as closing eyes and squinting, and smiling will introduce errors in the recognition process. Especially in complex environments, it is easy to cause a significant decrease in the recognition stability of the system.

[0003] In view of this, the present invention proposes an intelligent campus identity management system and method based on face recognition to solve the above problems. Summary of the Invention

[0004] To overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: An intelligent campus identity management system based on face recognition, comprising:

[0005] A data acquisition and processing module: collecting a sequence of face images and corresponding identity information, and performing dynamic feature tracking on the sequence of face images to obtain a sequence of feature marker maps;

[0006] A skin analysis module: performing skin state analysis on the sequence of feature marker maps to obtain a skin state feature set;

[0007] An expression analysis module: including a modality analysis unit and a modality compensation unit. The modality analysis unit performs expression analysis on the skin state feature set to obtain static features and micro-expression features, and the modality compensation unit performs micro-deformation compensation on the micro-expression features to obtain micro-motion compensation features;

[0008] A model construction module: constructing an identity recognition model based on the compensation features, static features, and identity information, and realizing accurate identification of personnel based on the identity recognition model.

[0009] Further, the acquisition method of the face image sequence includes:

[0010] Using a high-resolution camera in cooperation with a high-speed imaging device, the camera is equipped with an automatic focusing function, the camera is set to collect at a high frame rate, the collection frame rate is 60fps or higher, for the personnel appearing within the camera's field of view, one minute is used as the collection time, and image collection is performed at a collection interval of one frame per second. All the collected images are sorted from earliest to latest based on the collection time to obtain a face image sequence.

[0011] Further, the method for dynamically tracking the features of the face image sequence includes:

[0012] Preset standard pixel coordinates, mark the coordinates of the pixels of each frame of the face image sequence based on the standard pixel coordinates to obtain pixel coordinates; use the first frame of the face image sequence as the starting tracking image, use the other images in the face image sequence except the first frame as feature exploration images, adopt a facial feature point detection algorithm to mark the serial numbers of the feature points on the starting tracking image to obtain a marked feature image; use the marked pixels in the marked feature image as contour feature points, preset a displacement threshold, the displacement threshold is a positive integer, use the pixels at the same position as the pixel coordinates of the contour feature points in the feature exploration image as the center of the circle, use the displacement threshold as the radius to draw a circle, use the pixels within the circle as feature trajectory pixels, perform feature evaluation on each feature trajectory pixel to obtain a feature evaluation value; use the feature trajectory pixel with the largest feature evaluation value as the contour pixel, and use the codes marked on the contour feature points to perform coding and marking on the contour pixels to obtain an exploration marked image. All the exploration marked images and the starting tracking image form a feature marked image sequence.

[0013] Further, the formula for performing feature evaluation on each feature trajectory pixel is: where E represents the feature evaluation value, I represents the pixel value of the contour feature point, IB represents the pixel value of the feature trajectory pixel, x represents the horizontal axis value of the pixel coordinates of the contour feature point, y represents the vertical axis value of the pixel coordinates of the contour feature point, a represents the horizontal axis value of the pixel coordinates of the feature trajectory pixel, b represents the vertical axis value of the pixel coordinates of the feature trajectory pixel, and r represents the displacement threshold.

[0014] Further, the method for analyzing the skin state of the feature marked image sequence includes:

[0015] Taking the marked pixels in each frame image of the feature - marked image sequence as pixel analysis points, presetting a filter bank, initializing the filtering size, filtering direction angle, filtering wavelength, filtering bandwidth, and filtering phase of each filter, performing texture evaluation on the pixel analysis points based on the filter bank to obtain skin texture values; presetting a gradient analysis window, taking the pixel analysis point as the center of the gradient analysis window, using a gradient operator to perform gradient analysis on the pixels within the gradient analysis window to obtain window pixel gradients, and using the maximum - minimum normalization algorithm to normalize the mean value of the window pixel gradients to obtain pixel elastic coefficients; the skin texture value, window pixel gradient, and pixel elastic coefficient of the pixel analysis point constitute a skin state description, performing a skin state mapping on the skin state description to obtain a skin state value; the skin state values of all pixel analysis points in the same image constitute a skin feature matrix, and all skin feature matrices constitute a skin state feature set.

[0016] Further, the method for performing expression analysis on the skin state feature set includes:

[0017] Taking each skin feature matrix in the skin state feature set as a group to be analyzed, using a clustering analysis algorithm to perform clustering analysis on the skin state values in the group to be analyzed to obtain skin state clusters; taking the skin state cluster with the largest mean value of skin state values as the state center, using the skin average state and average pixel coordinates of the state center as the evaluation center, where the skin average state is the mean value of the skin state values within the state center, and the average pixel coordinates are the mean values of the pixel coordinates corresponding to each skin state value within the state center; performing local deformation evaluation on the group to be analyzed based on the evaluation center to obtain a micro - motion deformation field; all micro - motion deformation fields of the same group to be analyzed constitute an expression description; presetting a deformation threshold, taking the expression description with the mean value of the micro - motion deformation field within the expression description greater than or equal to the deformation threshold as a micro - expression feature, and taking the expression description with the mean value of the micro - motion deformation field within the expression description less than the deformation threshold as a static feature.

[0018] Further, the formula for performing local deformation evaluation on the group to be analyzed is: where D(p) represents the micro - motion deformation field of the p - th skin state value in the group to be analyzed, m represents the p - th skin state value in the group to be analyzed, Org represents the skin average state of the state center, len represents the coordinate distance of the state center, the coordinate distance of the state center is the modulus value of the average pixel coordinates of the state center, τ represents a balance parameter, Set p represents the pixel coordinates corresponding to the p - th skin state value in the group to be analyzed, o represents the average pixel coordinates of the state center, ρ represents a spatial parameter, and Dic() represents a distance measurement function.

[0019] Further, the method for performing micro - deformation compensation on the micro - expression feature includes:

[0020] Collect standard face images, use the left eye corner, right eye corner, left mouth corner, right mouth corner, and the center of the nose tip in the standard face image as deformation base points, mark the coordinates of the standard face image with standard pixel coordinates, record the pixel coordinates of the deformation base points, all deformation base points form a base point set, use each group of micro-expression features as the group to be compensated, and perform compensation evaluation on the group to be compensated based on the base point set to obtain micro-motion compensation; the micro-motion compensations corresponding to the micro-motion deformation fields within the same group to be compensated form a micro-motion compensation vector, and all micro-motion compensation vectors form micro-motion compensation features.

[0021] Further, the formula for performing compensation evaluation on the group to be compensated is:

[0022] Among them, CH z represents the micro-motion compensation of the z-th micro-motion deformation field in the group to be compensated, Val z represents the z-th micro-motion deformation field in the group to be compensated, U represents the size of the base point set, wei g represents the deformation weight of the g-th deformation base point in the base point set, Dic() represents the distance measurement function, VD z represents the pixel coordinates of the pixels corresponding to the z-th micro-motion deformation field in the group to be compensated, Drp g represents the pixel coordinates of the g-th deformation base point in the base point set, h represents the gray control parameter, represents the gray gradient of the pixel corresponding to the g-th deformation base point, represents the gray gradient of the pixels corresponding to the z-th micro-motion deformation field in the group to be compensated, represents the micro-motion gradient of the z-th micro-motion deformation field in the group to be compensated.

[0023] A smart campus identity management method based on face recognition includes:

[0024] S1. Collect a sequence of face images and corresponding identity information, perform dynamic feature tracking on the sequence of face images, and obtain a sequence of feature marker maps;

[0025] S2. Perform skin state analysis on the sequence of feature marker maps to obtain a skin state feature set;

[0026] S3. Perform expression analysis on the skin state feature set to obtain static features and micro-expression features, perform micro-deformation compensation on the micro-expression features, and obtain micro-motion compensation features;

[0027] S4. Construct an identity recognition model based on the compensation features, static features, and identity information, and achieve accurate identification of personnel based on the identity recognition model.

[0028] The technical effects and advantages of the smart campus identity management system and method based on face recognition of the present invention:

[0029] By dynamically tracking the facial image sequence, the present invention solves the deficiency of traditional static feature point detection methods in terms of feature point drift, realizes the accurate evaluation of the feature trajectory pixels of the facial image sequence, thereby ensuring the coherence and accuracy of the feature point trajectory; by analyzing the skin state of the feature marker map sequence, it can capture temporary changes such as facial swelling or eye bags caused by factors such as fatigue and staying up late, enabling the system to maintain a high recognition accuracy when dealing with non-persistent facial changes; through the expression analysis module, the system can recognize emotional expressions such as subtle eye movements or mouth corner changes, effectively reducing recognition errors caused by expression changes and enhancing the stability of the system in dynamic expressions and complex scenarios; by constructing an identity recognition model, it not only enhances the recognition ability of the system, but also improves the robustness and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 FIG. is a schematic diagram of a smart campus identity management system based on face recognition according to the present invention;

[0031] Figure 2 FIG. is a schematic diagram of a smart campus identity management method based on face recognition according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0033] Embodiment 1

[0034] Please refer to Figure 1 As shown, a smart campus identity management system based on face recognition in this embodiment includes:

[0035] Data acquisition and processing module: Collect facial image sequences and corresponding identity information, dynamically track the facial image sequences, and obtain a sequence of feature marker maps;

[0036] Skin analysis module: Analyze the skin state of the sequence of feature marker maps to obtain a skin state feature set;

[0037] Expression analysis module: including a mood analysis unit and a mood compensation unit. The mood analysis unit analyzes the skin state feature set for expressions to obtain static features and micro-expression features. The mood compensation unit performs micro-deformation compensation on the micro-expression features to obtain micro-motion compensation features;

[0038] Model construction module: construct an identity recognition model based on compensation features, static features and identity information, and achieve accurate identification of personnel based on the identity recognition model;

[0039] Each module is connected by wired and / or wireless means to achieve data transmission between modules;

[0040] The acquisition methods of the face image sequence include:

[0041] Use a high-resolution camera in combination with a high-speed imaging device to ensure that fast-changing facial expressions and postures can be captured. The camera is set to high-frame-rate acquisition, and the acquisition frame rate is 60fps or higher to ensure that sufficient data volume can be captured in fast dynamic scenarios. The camera is equipped with an autofocus function to ensure that even during movement, the camera can always focus clearly, ensuring that facial details are not distorted. For the personnel appearing within the camera's field of view, one minute is used as the acquisition time, and image acquisition is performed at an acquisition interval of one frame per second. All the acquired images are sorted from earliest to latest based on the acquisition time to obtain a face image sequence. All the obtained face image data have obtained the consent of the relevant personnel and comply with the provisions of relevant laws and policies.

[0042] In the case of rapid changes in face postures (such as raising the head, lowering the head, turning the head) or dynamic scenarios (such as continuous speaking, laughing), facial feature points are prone to drift. Traditional static feature point detection methods are difficult to track the changes of feature points in real time. Dynamic feature tracking of the face image sequence solves the problem of feature point drift caused by rapid posture changes and facial expression movements, ensuring the coherence and accuracy of the feature point trajectory, thereby providing a stable input for subsequent processing. Specifically:

[0043] Preset standard pixel coordinates, where both the horizontal and vertical coordinates in the standard pixel coordinates use one pixel as the coordinate scale; based on the standard pixel coordinates, mark the coordinates of the pixels in each frame of the face image sequence. Use the lower left corner of each frame of the face image sequence as the mapping point of the coordinate origin in the standard pixel coordinates. For each pixel increase to the right, the value of the horizontal axis of the coordinate is incremented by one, and for each pixel increase to the left, the value of the vertical axis of the coordinate is incremented by one to obtain the pixel coordinates; use the first frame of the face image sequence as the starting tracking image, and use the other images in the face image sequence except the first frame as the feature exploration images. Use the facial feature point detection algorithm to mark the serial numbers of the feature points on the starting tracking image to obtain the marked feature image. Common facial feature point detection algorithms include the Dlib 68-point facial marking method and the OpenCV facial feature point detection algorithm. The marked feature image contains the marks of the important part points (i.e., pixel points) of the face contour, and the marks are attached with the encoding of the feature point pixels to distinguish the feature points; use the marked pixels in the marked feature image as the contour feature points, preset a displacement threshold, and the displacement threshold is an integer greater than zero. Use the pixels at the same position as the pixel coordinates of the contour feature points in the feature exploration image as the center of the circle, and the displacement threshold as the radius to draw a circle. Use the pixels within the circle as the feature track pixels, and perform feature evaluation on each feature track pixel. The formula for performing feature evaluation on each feature track pixel is:

[0044] where E represents the feature evaluation value, I represents the pixel value of the contour feature point, IB represents the pixel value of the feature track pixel, x represents the horizontal axis value of the pixel coordinate of the contour feature point, y represents the vertical axis value of the pixel coordinate of the contour feature point, a represents the horizontal axis value of the pixel coordinate of the feature track pixel, b represents the vertical axis value of the pixel coordinate of the feature track pixel, and r represents the displacement threshold; use the feature track pixel with the largest feature evaluation value as the contour pixel, and use the encoding marked on the contour feature point to encode and mark the contour pixel to obtain the exploration marked image. All the exploration marked images and the starting tracking image form the feature marked image sequence.

[0045] Traditional face recognition systems have poor recognition effects when dealing with temporary facial changes such as facial swelling and eye bags caused by lack of sleep, staying up late, fatigue, etc. The reason is that facial swelling and eye bags will cause distortion of facial features, making it difficult for the system to match accurate identity information. By analyzing the skin state of the feature marked image sequence, the skin feature changes caused by facial swelling or other physiological changes can be captured, enabling the system to adapt to these temporary changes rather than relying on the original static image features. Specifically:

[0046] Taking the marked pixels in each frame image of the feature - marked image sequence as pixel analysis points, presetting a filter bank, initializing the filtering size, filtering direction angle, filtering wavelength, filtering bandwidth, and filtering phase of each filter, and performing texture evaluation on the pixel analysis points based on the filter bank. The formula for performing texture evaluation on the pixel analysis points is: where T represents the skin texture value, S represents the pixel coordinates of the pixel analysis point, G c represents the c - th filter, * represents the filtering convolution operation, N represents the number of filters, which is used to multiply and sum the filter with the local image region divided by taking the pixel coordinates of the pixel analysis point as the center and the filtering size as the local window, and output the texture response in a specific direction; common filters include Gabor filters. The filters can extract the directional features of the texture, such as the fineness and texture direction of the skin;

[0047] Presetting a gradient analysis window, taking the pixel analysis point as the center of the gradient analysis window, using a gradient operator to perform gradient analysis on the pixels within the gradient analysis window to obtain the window pixel gradient, and normalizing the mean value of the window pixel gradient using the maximum - minimum normalization algorithm to obtain the pixel elasticity coefficient. Common gradient operators include Sobel operators and Prewitt operators; the skin texture value, window pixel gradient, and pixel elasticity coefficient of the pixel analysis point constitute the skin state description. Performing skin state mapping on the skin state description, the formula for performing skin state mapping on the skin state description is: M = tanh(W×Cri + μ); where M represents the skin state value, W represents the state weight matrix, Cri represents the skin state description, μ represents the bias parameter used to adjust the balance of the mapping, and tanh() represents the hyperbolic tangent function; the skin state values of all pixel analysis points in the same image constitute the skin feature matrix, and all skin feature matrices constitute the skin state feature set.

[0048] Micro - expressions (such as subtle eye movements, slight upward curvature of the mouth corners, etc.) are relatively concealed and rapid changes for a face recognition system. Traditional recognition technologies are difficult to accurately capture these subtle facial changes. Micro - expressions reflect the emotional state and physiological changes of an individual, and these subtle changes will affect the accuracy of the recognition system; by performing expression analysis on the skin state feature set to extract micro - expression features, especially when the user's expression changes greatly, the system can better adapt to the rapid dynamic changes of the face. Specifically:

[0049] Taking each skin feature matrix in the skin state feature set as the group to be analyzed, a clustering analysis algorithm is used to perform clustering analysis on the skin state values in the group to be analyzed to obtain skin state clusters. Commonly used clustering analysis algorithms include the K-Means clustering algorithm and the hierarchical clustering algorithm; taking the skin state cluster with the largest mean value of skin state values as the state center, and using the skin average state and average pixel coordinates of the state center as the evaluation center. The skin average state is the mean value of the skin state values within the state center, and the average pixel coordinates are the mean values of the pixel coordinates corresponding to each skin state value within the state center; based on the evaluation center, a local deformation evaluation is performed on the group to be analyzed. The formula for performing local deformation evaluation is:

[0050] where D(p) represents the micro-deformation field of the p-th skin state value in the group to be analyzed, m represents the p-th skin state value in the group to be analyzed, Org represents the skin average state of the state center, len represents the coordinate distance of the state center, the coordinate distance of the state center is the modulus value of the average pixel coordinates of the state center, τ represents the balance parameter used to control the balance between the skin state value difference and the spatial distance, Set p represents the pixel coordinates corresponding to the p-th skin state value in the group to be analyzed, o represents the average pixel coordinates of the state center, ρ represents the spatial parameter used to balance the influence of the spatial distance on the micro-deformation field, and Dic() represents the distance measurement function. Common distance measurement functions include the Euclidean distance function and the Manhattan distance function; all the micro-deformation fields of the same group to be analyzed constitute the expression description; a deformation threshold is preset, and the deformation threshold is set by those skilled in the art according to the actual situation. The expression description with the mean value of the micro-deformation fields within the expression description greater than or equal to the deformation threshold is used as the micro-expression feature, and the expression description with the mean value of the micro-deformation fields within the expression description less than the deformation threshold is used as the static feature.

[0051] The rapid deformation of facial muscles, expression changes, etc. will cause the drift of the feature point positions, affecting the stability and accuracy of facial recognition, especially in the case of expression changes or dynamic scenes, the geometric shape of the face changes; by performing micro-deformation compensation on the micro-expression features, the changes in the facial shape can be adjusted at multiple levels, reducing the influence of expression changes on the recognition result, enabling the system to stably process the facial deformation in a dynamic environment, and being able to more accurately reflect the actual facial state. Specifically:

[0052] Collect standard face images. The standard face images are grayscale images that have undergone face detection and alignment processing, with the position, scale, and orientation of the face unified, reducing the interference of color on recognition and retaining the shape and structural information of the face. The standard face images are consistent in size and resolution with the images in the face image sequence; use the left eye corner, right eye corner, left mouth corner, right mouth corner, and the center of the nose tip in the standard face image as deformation base points, mark the coordinates of the standard face image with standard pixel coordinates, record the pixel coordinates of the deformation base points, and all the deformation base points form a base point set. Use each group of micro-expression features as the group to be compensated, and perform compensation evaluation on the group to be compensated based on the base point set. The formula for compensating and evaluating the group to be compensated is: Among them, CH z represents the micro-motion compensation of the z-th micro-motion deformation field in the group to be compensated, Val z represents the z-th micro-motion deformation field in the group to be compensated, U represents the size of the base point set, wei g represents the deformation weight of the g-th deformation base point in the base point set, Dic() represents the distance measurement function, and common distance measurement functions include the Euclidean distance function and the Manhattan distance function. VD z represents the pixel coordinates of the pixel corresponding to the z-th micro-motion deformation field in the group to be compensated, Drp g represents the pixel coordinates of the g-th deformation base point in the base point set, h represents the gray control parameter, which is used to control the sensitivity of the gray contrast, represents the gray gradient of the pixel corresponding to the g-th deformation base point, represents the gray gradient of the pixel corresponding to the z-th micro-motion deformation field in the group to be compensated, represents the micro-motion gradient of the z-th micro-motion deformation field in the group to be compensated; the micro-motion compensations corresponding to the micro-motion deformation fields within the same group to be compensated form a micro-motion compensation vector, and all the micro-motion compensation vectors form micro-motion compensation features.

[0053] Based on the compensation features, static features, and identity information, use the CNN model as the initial model of the identity recognition model, use the compensation features, static features, and identity information as training data, use the training data as the training sample set, train the CNN model with the training sample set, use the compensation features, static features, and identity information as the input data of the identity recognition model, and use the predicted identity label as the output data of the identity recognition model; use minimizing the error between the actual identity information and the identity label predicted by the identity recognition model as the training objective, use the recall rate function as the loss function of the identity recognition model, and stop training to obtain the identity recognition model when the loss function converges.

[0054] In this embodiment, by performing dynamic feature tracking on the face image sequence, the deficiencies of traditional static feature point detection methods in terms of feature point drift are addressed, and accurate evaluation of the feature trajectory pixels of the face image sequence is achieved, thus ensuring the coherence and accuracy of the feature point trajectory; by performing skin state analysis on the feature marker map sequence, temporary changes such as facial swelling or eye bags caused by factors such as fatigue and staying up late can be captured, enabling the system to maintain a high recognition accuracy when dealing with non-persistent facial changes; through the expression analysis module, the system can recognize emotional expressions such as subtle eye movements or mouth corner changes, effectively reducing recognition errors caused by expression changes and enhancing the stability of the system in dynamic expressions and complex scenarios; by constructing an identity recognition model, not only the recognition ability of the system is enhanced, but also the robustness and adaptability of the system are improved.

[0055] Embodiment 2

[0056] Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A smart campus identity management method based on face recognition is provided, including:

[0057] S1. Collect a face image sequence and corresponding identity information, perform dynamic feature tracking on the face image sequence, and obtain a feature marker map sequence;

[0058] S2. Perform skin state analysis on the feature marker map sequence to obtain a skin state feature set;

[0059] S3. Perform expression analysis on the skin state feature set to obtain static features and micro-expression features, and perform micro-deformation compensation on the micro-expression features to obtain micro-motion compensation features;

[0060] S4. Construct an identity recognition model based on the compensation features, static features, and identity information, and achieve accurate identification of personnel based on the identity recognition model.

[0061] Embodiment 3

[0062] This embodiment publicly provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the operation mode of the above-provided smart campus identity management method based on face recognition.

[0063] Since the electronic device introduced in this embodiment is the electronic device adopted for implementing a smart campus identity management method based on face recognition in the embodiments of the present application, based on the method for a smart campus identity management method based on face recognition introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiments of the present application will not be introduced in detail here. As long as those skilled in the art implement the electronic device adopted for a smart campus identity management method based on face recognition in the embodiments of the present application, it falls within the scope of protection of the present application.

[0064] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain a formula that is closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0065] The above are only the preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the scope of protection of the present invention. It should be noted that for ordinary technical users in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the scope of protection of the present invention.

Claims

1. A smart campus identity management system based on face recognition, characterized in that: include: Data acquisition and processing module: collects facial image sequences and corresponding identity information, performs dynamic feature tracking on facial image sequences, and obtains feature marker image sequences; Skin analysis module: performs skin condition analysis on the feature marker image sequence to obtain a skin condition feature set; Expression analysis module: including a modality analysis unit and a modality compensation unit. The modality analysis unit performs expression analysis on the skin state feature set to obtain static features and micro-expression features. The modality compensation unit performs micro-deformation compensation on the micro-expression features to obtain micro-motion compensation features. Model building module: builds an identity recognition model based on compensation features, static features and identity information, and realizes accurate identification of personnel based on the identity recognition model; The method of performing skin condition analysis on the characteristic marker sequence includes: The marked pixels in each frame image in the feature marking image sequence are used as pixel analysis points, a filter group is preset, the filter size, filter direction angle, filter wavelength, filter bandwidth and filter phase of each filter are initialized, and texture evaluation is performed on the pixel analysis point based on the filter group to obtain skin texture value; a gradient analysis window is preset, the pixel analysis point is used as the center of the gradient analysis window, a gradient operator is used to perform gradient analysis on the pixels in the gradient analysis window to obtain the window pixel gradient, and the maximum and minimum normalization algorithm is used to normalize the mean of the window pixel gradient to obtain the pixel elasticity coefficient; the skin texture value of the pixel analysis point, the window pixel gradient and the pixel elasticity coefficient constitute a skin state description, and the skin state description is subjected to skin state mapping to obtain the skin state value; the skin state values ​​of all pixel analysis points in the same image constitute a skin feature matrix, and all skin feature matrices constitute a skin state feature set; The method of performing expression analysis on the skin state feature set includes: Each skin feature matrix in the skin state feature set is taken as the group to be analyzed, and the skin state values ​​in the group to be analyzed are clustered by a cluster analysis algorithm to obtain skin state clusters; the skin state cluster with the largest mean of skin state values ​​is taken as the state center, and the skin homogeneity and average pixel coordinates of the state center are taken as the evaluation center, the skin homogeneity is the mean of the skin state values ​​in the state center, and the average pixel coordinates are the mean of the pixel coordinates corresponding to each skin state value in the state center; based on the evaluation center, a local deformation evaluation is performed on the group to be analyzed to obtain a micro-motion deformation field; all micro-motion deformation fields of the same group to be analyzed constitute an expression description; a deformation threshold is preset, and the expression description whose mean of the micro-motion deformation field in the expression description is greater than or equal to the deformation threshold is taken as a micro-expression feature, and the expression description whose mean of the micro-motion deformation field in the expression description is less than the deformation threshold is taken as a static feature.

2. The smart campus identity management system based on face recognition according to claim 1 is characterized in that: The acquisition method of the face image sequence includes: A high-resolution camera is used in conjunction with a high-speed camera. The camera is equipped with an automatic focus function and is set to high frame rate acquisition. The acquisition frame rate is 60fps or higher. The acquisition time is one minute for people who appear in the camera's field of view, and image acquisition is performed at an interval of one frame per second. All acquired images are sorted from early to late based on the acquisition time to obtain a facial image sequence.

3. The smart campus identity management system based on face recognition according to claim 2 is characterized in that: The method of performing dynamic feature tracking on a facial image sequence includes: Preset standard pixel coordinates, and mark the coordinates of the pixels of each frame in the face image sequence based on the standard pixel coordinates to obtain pixel coordinates; use the first frame in the face image sequence as the starting tracking image, and use the other images in the face image sequence except the first frame as the feature exploration image, and use the facial feature point detection algorithm to mark the feature point serial number of the starting tracking image to obtain a marked feature image; use the marked pixels in the marked feature image as contour feature points, preset a displacement threshold, and the displacement threshold is an integer greater than zero, use the pixels at the same position as the pixel coordinates of the contour feature point in the feature exploration image as the center of the circle, and use the displacement threshold as the radius to draw a circle, use the pixels within the circle as feature track pixels, and perform feature evaluation on each feature track pixel to obtain a feature evaluation value; use the feature track pixel with the largest feature evaluation value as the contour pixel, and encode and mark the contour pixel with the code marked in the contour feature point to obtain a search mark image, and all search mark images and the starting tracking image constitute a feature mark image sequence.

4. The smart campus identity management system based on face recognition according to claim 3 is characterized in that: The formula for evaluating the feature of each feature track pixel is: Among them, E represents the feature evaluation value, I represents the pixel value of the contour feature point, IB represents the pixel value of the feature trajectory pixel, x represents the horizontal axis value of the pixel coordinate of the contour feature point, y represents the vertical axis value of the pixel coordinate of the contour feature point, a represents the horizontal axis value of the pixel coordinate of the feature trajectory pixel, b represents the vertical axis value of the pixel coordinate of the feature trajectory pixel, and r represents the displacement threshold.

5. The smart campus identity management system based on face recognition according to claim 4 is characterized in that: The formula for evaluating the local deformation of the group to be analyzed is: Where D(p) represents the micro-deformation field of the p-th skin state value in the group to be analyzed, m represents the p-th skin state value in the group to be analyzed, Org represents the skin homogeneity of the state center, len represents the coordinate distance of the state center, the coordinate distance of the state center is the modulus of the average pixel coordinate of the state center, τ represents the balance parameter, Set p represents the pixel coordinates corresponding to the pth skin state value in the group to be analyzed, o represents the average pixel coordinates of the state center, ρ represents the spatial parameter, and Dic() represents the distance measurement function.

6. The smart campus identity management system based on face recognition according to claim 5 is characterized in that: The method of performing micro deformation compensation on micro expression features includes: A standard face image is collected, and the left eye corner, right eye corner, left mouth corner, right mouth corner and nose tip center in the standard face image are used as deformation base points. The standard face image is marked with standard pixel coordinates, and the pixel coordinates of the deformation base points are recorded. All deformation base points constitute a base point set. Each group of micro-expression features is used as a group to be compensated. The group to be compensated is evaluated based on the base point set to obtain micro-motion compensation. The micro-motion compensation corresponding to the micro-motion deformation field in the same group to be compensated constitutes a micro-motion compensation vector, and all micro-motion compensation vectors constitute micro-motion compensation features.

7. The smart campus identity management system based on face recognition according to claim 6 is characterized in that: The formula for evaluating compensation for the group to be compensated is: Among them, CH z Represents the micro-motion compensation of the zth micro-motion deformation field in the group to be compensated, Val z represents the zth micro-deformation field in the group to be compensated, U represents the size of the base point set, wei g represents the deformation weight of the g-th deformation base point in the base point set, Dic() represents the distance measurement function, VD z represents the pixel coordinates of the pixel corresponding to the zth micro-deformation field in the group to be compensated, Drp g represents the pixel coordinates of the gth deformation base point in the base point set, h represents the grayscale control parameter, Represents the grayscale gradient of the pixel corresponding to the g-th deformation base point, represents the grayscale gradient of the pixel corresponding to the zth micro-motion deformation field in the group to be compensated, Represents the micromotion gradient of the zth micromotion deformation field in the group to be compensated.

8. A smart campus identity management method based on face recognition, which is implemented based on the smart campus identity management system based on face recognition according to any one of claims 1 to 7, characterized in that: include: S1, collecting a facial image sequence and corresponding identity information, performing dynamic feature tracking on the facial image sequence, and obtaining a feature marker sequence; S2, performing skin condition analysis on the feature marker image sequence to obtain a skin condition feature set; S3, performing expression analysis on the skin state feature set to obtain static features and micro-expression features, performing micro-deformation compensation on the micro-expression features to obtain micro-motion compensation features; S4. Construct an identity recognition model based on compensation features, static features and identity information, and achieve accurate identification of personnel based on the identity recognition model.

Images