Check-in method based on facial recognition

Abstract

This invention provides a face recognition-based check-in method. It reads facial images of a target person captured by several cameras, along with their capture time information. Based on the capture time information, background content, and facial features, the target person's image is sequentially filtered by capture time, location, and facial features. Then, from the capture time information of the final filtered facial image, the check-in start and end times are determined, thus judging the validity of the check-in. This check-in method utilizes distributed cameras to centrally capture facial images of the target person, allowing staff to easily access the cameras anytime, anywhere. The combination of capture time information, background content, and facial features ensures that only facial images meeting the specified criteria are retained, effectively preventing fraudulent check-in attempts.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent attendance management, and in particular to an attendance method based on facial recognition. Background Technology

[0002] Currently, various check-in methods, such as location-based check-in, QR code check-in, gesture check-in, and biometric check-in, can meet different work attendance needs. However, these methods all have some drawbacks, making it inevitable that there will be cases of fraudulent check-in, reducing the authenticity of check-in information, and making it impossible to accurately obtain employees' check-in times. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a face recognition-based check-in method. This method reads facial images of a target person captured by several cameras, along with their capture time information. Based on the capture time information, background content, and facial features, the target person's image is sequentially filtered by capture time, location, and facial features. The capture time information of the final filtered image is used to determine the start and end times of the check-in, thereby judging the validity of the check-in. Based on the judgment, a corresponding notification message is sent to the target person's terminal device. This check-in method utilizes distributed cameras to centrally capture facial images of the target person, allowing staff to easily access the cameras anytime, anywhere. The combination of capture time information, background content, and facial features filtering retains facial images that meet the relevant conditions, effectively preventing fraudulent check-ins and improving the authenticity of check-in information.

[0004] This invention provides a check-in method based on facial recognition, which includes the following steps:

[0005] Step S1: Read the facial images of the target person captured by each of the several cameras and their capture time information; based on the capture time information, determine whether the capture time of each facial image of the target person meets the preset time condition.

[0006] Step S2: Extract the background content of the images from all the target personnel face images that meet the preset time conditions, and filter out all the target personnel face images that meet the preset location conditions based on the background content.

[0007] Step S3: Extract facial features of the target personnel from all facial images that meet the preset location conditions, and filter out all facial images of the target personnel that meet the preset personnel identity conditions based on the facial features.

[0008] Step S4: Based on the shooting time information of the facial images of all target personnel that meet the preset personnel identity conditions, obtain the check-in start time and check-in end time of the target personnel, and thus determine whether the check-in of the target personnel is valid; then, based on the above judgment result, send the corresponding notification message to the target personnel's terminal device.

[0009] Furthermore, in step S1, before reading the target person's facial image and its shooting time information, the following steps are also included:

[0010] The system instructs the target person's terminal device to connect to a camera, and uses the terminal device to control the camera to take pictures of the target person and obtain an image of the target person's face.

[0011] Upload the captured facial images of the target personnel and their shooting time information to the check-in management platform;

[0012] Based on the chronological order in which the facial images of each target person were captured, the attendance management platform is instructed to arrange all uploaded facial images of the target persons in sequence to form an image set sequence.

[0013] Furthermore, in step S1, after instructing the target person's terminal device to connect to the camera, and using the terminal device to control the camera to capture an image of the target person's face, the process further includes:

[0014] Several supplementary lights are installed on each camera, so that all the supplementary lights are evenly distributed around the camera, and the direction of the light illumination of the supplementary lights is adjusted to be the same as the shooting direction of the camera. Based on the RGB values ​​and brightness distribution of the captured facial image of the target person, corresponding supplementary lighting operations are performed on the target person, and new facial images of the target person are captured again during the above supplementary lighting operations. The process is as follows:

[0015] Step S101: Using the following formula (1), obtain the color distribution value of the target person's face image based on the RGB values ​​of the captured image.

[0016]

[0017] In the above formula (1), D represents the color distribution value of the target person's face image; [R(i,j),G(i,j),B(i,j)] represents the RGB value of the pixel in the i-th row and j-th column of the target person's face image; m represents the number of pixels in each row of the target person's face image; n represents the number of pixels in each column of the target person's face image. This means substituting the values ​​of i from 1 to n and j from 1 to m into the parentheses to get the maximum value inside the parentheses; || means to calculate the absolute value;

[0018] Step S102: After performing pixel grayscale processing on the target person's face image, the brightness distribution value of the target person's face image is obtained using the following formula (2) based on the grayscale values ​​of the target person's face image after pixel grayscale processing.

[0019]

[0020] In the above formula (2), G represents the brightness distribution value of the target person's face image; H(i,j) represents the gray value of the pixel in the i-th row and j-th column after pixel grayscale processing of the target person's face image; H M This represents the maximum grayscale value, which is 255.

[0021] Step S103: Using the following formula (3), based on the color distribution and brightness distribution values ​​of the target person's facial image and the total number of fill lights installed on the camera, control the number of fill lights installed on the camera to be lit.

[0022]

[0023] In the above formula (3), q represents the number of fill lights installed on the camera; Q represents the total number of fill lights installed on the camera;

[0024] Based on the number q of fill lights installed on the camera, the fill lights are controlled to cycle through the lighting process. The process is as follows: starting from the fill light installed at the top of the camera, q fill lights are lit clockwise. Then, after a certain period of time, the (q+1)th fill light clockwise is lit and the first fill light is turned off. This process is repeated to cycle through the clockwise lighting.

[0025] Furthermore, in step S1, determining whether the shooting time of each target person's facial image meets the preset time condition based on the shooting time information specifically includes:

[0026] The capture time of each target person's facial image is obtained, and the capture time is compared with a preset work start time range. If the capture time is within the preset work start time range, the corresponding target person's facial image is determined to meet the preset time condition; otherwise, the corresponding target person's facial image is determined not to meet the preset time condition.

[0027] Furthermore, in step S2, the background content of the image extracted from all target personnel facial images that meet the preset time conditions specifically includes:

[0028] Identify the contour boundary between the target person and the background environment in the image from all facial images of the target person that meet the preset time conditions;

[0029] Based on the outline boundary line, the background content of the image is extracted from the facial image of the target person.

[0030] Furthermore, in step S2, filtering out all target personnel facial images that meet the preset location conditions based on the background image content specifically includes:

[0031] Information on the types of background objects present in the background environment of the image frame containing the target person's face is obtained from the background image content.

[0032] Based on the background object type information, determine the actual shooting location of the target person's face image, and thereby determine whether the actual shooting location belongs to the preset shooting location;

[0033] If the location is specified, the corresponding target person's facial image is determined to meet the preset location conditions; otherwise, the corresponding target person's facial image is determined not to meet the preset location conditions.

[0034] Furthermore, in step S3, extracting facial features from all target personnel facial images that meet the preset location conditions specifically includes:

[0035] Facial features of individuals are extracted from facial images of all target individuals that meet the predetermined location conditions.

[0036] Furthermore, in step S3, filtering out all target person facial images that meet the preset person identity conditions based on the person's facial features specifically includes:

[0037] The facial features of the person are compared with a preset database of facial features of staff members to determine the identity information of the target person corresponding to each target person's facial image.

[0038] Based on the target personnel's identity information, facial images of target personnel that were not taken by staff members are removed, while facial images of target personnel that were taken by staff members are retained, thereby filtering out all facial images of target personnel that meet the preset personnel identity conditions.

[0039] Furthermore, in step S4, based on the shooting time information corresponding to the facial images of all target personnel who meet the preset personnel identity conditions, the start and end times of the target personnel's check-in are obtained, thereby determining whether the target personnel's check-in is valid; then, based on the above determination result, a corresponding notification message is sent to the target personnel's terminal device, specifically including:

[0040] From the shooting time information corresponding to the facial images of all target personnel who meet the preset personnel identity conditions, the earliest and latest shooting times of the same target personnel are selected and used as the start and end times of the target personnel's check-in.

[0041] Determine whether the time difference between the check-in end time and the check-in start time is greater than or equal to a preset time range threshold. If yes, the check-in of the target person is deemed valid; otherwise, the check-in of the target person is deemed invalid.

[0042] If the target person's clock-in is invalid, a clock-in invalidation notification message will be sent to the target person's terminal device.

[0043] Compared to existing technologies, this face recognition-based check-in method reads facial images of the target person captured by several cameras, along with their capture time information. Based on the capture time information, background content, and facial features, the image is sequentially filtered by capture time, location, and facial features. The start and end times of the check-in are then determined from the capture time information of the final filtered facial images, thus judging the validity of the check-in. Based on the judgment, a corresponding notification message is sent to the target person's terminal device. This check-in method utilizes distributed cameras to centrally capture facial images of the target person, allowing staff to access the cameras anytime, anywhere. The combination of capture time information, background content, and facial features filtering retains facial images that meet the relevant conditions, effectively preventing fraudulent check-ins and improving the authenticity of the check-in information.

[0044] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0045] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1This is a flowchart illustrating the face recognition-based attendance method provided by the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] See Figure 1 This is a flowchart illustrating the face recognition-based attendance method provided in an embodiment of the present invention. The face recognition-based attendance method includes the following steps:

[0050] Step S1: Read the facial images of the target person captured by each of the several cameras and their capture time information; based on the capture time information, determine whether the capture time of each target person's facial image meets the preset time condition.

[0051] Step S2: Extract the background content of the images from all the target personnel face images that meet the preset time conditions, and filter out all the target personnel face images that meet the preset location conditions based on the background content.

[0052] Step S3: Extract facial features of the target personnel from all facial images that meet the preset location conditions, and filter out all facial images of the target personnel that meet the preset personnel identity conditions based on the facial features.

[0053] Step S4: Based on the shooting time information of the facial images of all target personnel that meet the preset personnel identity conditions, obtain the check-in start time and check-in end time of the target personnel, and thus determine whether the check-in of the target personnel is valid; then, based on the above judgment result, send the corresponding notification message to the target personnel's terminal device.

[0054] The beneficial effects of the above technical solution are as follows: This face recognition-based check-in method reads the facial images of the target personnel captured by several cameras and their capture time information. Based on the capture time information of the target personnel's face, the background content of the image, and the facial features of the personnel, the target personnel images are sequentially filtered by capture time, capture location, and facial features. Then, from the capture time information of the final filtered target personnel facial images, the check-in start time and check-in end time of the target personnel are obtained, thereby determining whether the check-in of the target personnel is valid. Based on the above determination result, a corresponding notification message is sent to the target personnel's terminal device. The above check-in method uses distributed cameras to centrally capture the facial images of the target personnel, which makes it convenient for staff to turn on the cameras anytime and anywhere. By combining the capture time information, the background content of the image, and the multiple filtering methods of facial features, the target personnel facial images that meet the corresponding conditions are retained. This can effectively prevent impersonation check-in and improve the authenticity of check-in information.

[0055] Preferably, in step S1, before reading the facial image of the target person and its shooting time information, the method further includes:

[0056] The system instructs the target person's terminal device to connect to a camera, and uses the terminal device to control the camera to take pictures of the target person and obtain an image of the target person's face.

[0057] Upload the captured facial images of the target personnel and their shooting time information to the check-in management platform;

[0058] Based on the chronological order in which the facial images of each target person were captured, the attendance management platform is instructed to arrange all uploaded facial images of the target persons in sequence to form an image set sequence.

[0059] The beneficial effects of the above technical solution are as follows: In practical work, cameras distributed in different locations can be connected via the Internet of Things (IoT) to achieve IoT control of multiple cameras. When a person approaches one of the cameras to take a picture, the person can instruct their terminal device to establish a wireless connection with the camera via Bluetooth or other means. This allows the person to indirectly control the camera to take a picture of them, and they can directly view the image of the person captured by the camera through their terminal device. Furthermore, the captured facial images of the person and their shooting time information are uploaded to the attendance management platform, which then arranges all uploaded facial images of the person in sequence to form an image set sequence. This improves the efficiency and regularity of arranging the facial images of the person.

[0060] Preferably, in step S1, after instructing the target person's terminal device to connect to the camera, and using the terminal device to control the camera to capture an image of the target person's face, the method further includes:

[0061] Several supplementary lights are installed on each camera, so that all the supplementary lights are evenly distributed around the camera, and the direction of the light illumination of the supplementary lights is adjusted to be the same as the shooting direction of the camera. Based on the RGB values ​​and brightness distribution of the captured facial image of the target person, corresponding supplementary lighting operations are performed on the target person, and new facial images of the target person are captured again during the above supplementary lighting operations. The process is as follows:

[0062] Step S101: Using the following formula (1), obtain the color distribution value of the target person's face image based on the RGB values ​​of the captured image.

[0063]

[0064] In the above formula (1), D represents the color distribution value of the target person's face image; [R(i,j),G(i,j),B(i,j)] represents the RGB value of the pixel in the i-th row and j-th column of the target person's face image; m represents the number of pixels in each row of the target person's face image; n represents the number of pixels in each column of the target person's face image. This means substituting the values ​​of i from 1 to n and j from 1 to m into the parentheses to get the maximum value inside the parentheses; || means to calculate the absolute value;

[0065] Step S102: After performing pixel grayscale processing on the target person's face image, the brightness distribution value of the target person's face image is obtained using the following formula (2) based on the grayscale values ​​of the target person's face image after pixel grayscale processing.

[0066]

[0067] In the above formula (2), G represents the brightness distribution value of the target person's face image; H(i,j) represents the gray value of the pixel in the i-th row and j-th column after pixel grayscale processing of the target person's face image; H M This represents the maximum grayscale value, which is 255.

[0068] Step S103: Using the following formula (3), based on the color distribution and brightness distribution values ​​of the target person's facial image and the total number of fill lights installed on the camera, control the number of fill lights installed on the camera to be lit.

[0069]

[0070] In the above formula (3), q represents the number of fill lights installed on the camera; Q represents the total number of fill lights installed on the camera;

[0071] Based on the number q of fill lights installed on the camera, the fill lights are controlled to cycle through the lighting process. The process is as follows: starting from the fill light installed at the top of the camera, q fill lights are lit clockwise. Then, after a certain period of time, the (q+1)th fill light clockwise is lit and the first fill light is turned off. This process is repeated to cycle through the clockwise lighting.

[0072] The beneficial effects of the above technical solution are as follows: Using the above formula (1), the color distribution value of the target person's face image is obtained based on the RGB value of the captured target person's face image, thereby indirectly judging from the color aspect whether the problem of color monotony is caused by the dark shooting environment; then using the above formula (2), the brightness distribution value of the target person's face image is obtained based on the gray value of the image after grayscale processing, thereby knowing the overall brightness distribution of the current environment, which is convenient for subsequent supplementary lighting of the environment; finally, using the above formula (3), the number of supplementary lights installed on the camera is controlled based on the color distribution value, brightness distribution value and the total number of supplementary lights installed on the camera, thereby performing reasonable and intelligent supplementary lighting. First, it can achieve the effect of energy saving. Second, by using the cyclic lighting method of supplementary lights, each supplementary light can work evenly, improving the lifespan of the system's supplementary lights.

[0073] Preferably, in step S1, determining whether the shooting time of each target person's facial image meets the preset time condition based on the shooting time information specifically includes:

[0074] The capture time of each target person's facial image is obtained, and the capture time is compared with a preset work start time range. If the capture time is within the preset work start time range, the corresponding target person's facial image is determined to meet the preset time condition; otherwise, the corresponding target person's facial image is determined not to meet the preset time condition.

[0075] The beneficial effects of the above technical solution are as follows: Since the preset work time range refers to the work clock-in time stipulated by the company, by using the above method, only the facial images of the target personnel whose shooting time is within the preset work time range are retained. This can effectively discard the facial images of the target personnel whose shooting time is not within the preset work time range, and avoid misjudgment of the clock-in duration in the future.

[0076] Preferably, in step S2, the background content of the image extracted from the facial images of all target personnel that meet the preset time conditions specifically includes:

[0077] Identify the contour boundary between the target person and the background environment in the image from all facial images of the target person that meet the preset time conditions;

[0078] Based on the outline boundary line, the background content of the image is extracted from the face image of the target person.

[0079] The beneficial effects of the above technical solution are as follows: through the above method, the background content of the target person’s face image can be accurately distinguished and identified from the target person’s face image, thus accurately determining the existence of objects in the background image.

[0080] Preferably, in step S2, selecting all target personnel facial images that meet the preset location conditions based on the background image content specifically includes:

[0081] The background image of the target person's face is identified from the background content, revealing the types of background objects present in the background environment.

[0082] Based on the background object type information, determine the actual shooting location of the target person's face image, and thus determine whether the actual shooting location belongs to the preset shooting location;

[0083] If the location is specified, the corresponding target person's facial image is determined to meet the preset location conditions; otherwise, the corresponding target person's facial image is determined not to meet the preset location conditions.

[0084] The beneficial effects of the above technical solution are as follows: by using the above method, based on the type of background objects present in the background content of the target person's facial image, the actual shooting location of the target person's facial image can be accurately determined, thereby further determining whether the actual shooting location is the pre-set check-in location.

[0085] Preferably, in step S3, extracting facial features from all target personnel facial images that meet the preset location conditions specifically includes:

[0086] Facial features of individuals are extracted from facial images of all target individuals that meet the predetermined location conditions.

[0087] The beneficial effects of the above technical solution are: the facial features of the target personnel are extracted from the facial images of all target personnel that meet the predetermined location conditions, and the facial features of the target personnel can be used as the unique authentication standard for the identity information of the target personnel.

[0088] Preferably, in step S3, filtering out all target person facial images that meet the preset person identity criteria based on the person's facial features specifically includes:

[0089] The facial features of the person are compared with the preset database of facial features of staff members to determine the identity information of the target person corresponding to each target person's facial image;

[0090] Based on the target person's identity information, facial images of the target person that were not taken by staff are removed, while facial images of the target person that were taken by staff are retained, thereby filtering out all facial images of the target person that meet the preset personnel identity conditions.

[0091] The beneficial effects of the above technical solution are as follows: by removing facial images of target personnel that are not taken by staff members and retaining facial images of target personnel that are taken by staff members, it is possible to effectively prevent the occurrence of impersonation during check-in.

[0092] Preferably, in step S4, based on the shooting time information corresponding to the facial images of all target personnel that meet the preset personnel identity conditions, the start time and end time of the target personnel's check-in are obtained, thereby determining whether the target personnel's check-in is valid; then, based on the above determination result, sending a corresponding notification message to the target personnel's terminal device specifically includes:

[0093] From the shooting time information corresponding to the facial images of all target personnel who meet the preset personnel identity conditions, the earliest and latest shooting times of the same target personnel are selected and used as the start and end times of the target personnel's check-in.

[0094] Determine whether the time difference between the end time and start time of the check-in is greater than or equal to a preset time range threshold. If yes, the check-in of the target person is deemed valid; otherwise, the check-in of the target person is deemed invalid.

[0095] If the target person's clock-in is invalid, a clock-in invalidation notification message will be sent to the target person's terminal device.

[0096] The beneficial effects of the above technical solution are as follows: by using the time difference between the end time and the start time of the check-in, the duration of the target employee's work is measured to determine whether the check-in is valid, thereby accurately judging whether the check-in is valid and then sending a corresponding notification message to the target employee's terminal device to ensure that the target employee is aware of their check-in status in real time.

[0097] As can be seen from the above embodiments, this face recognition-based check-in method reads the target person's facial images and their shooting time information captured by several cameras. Based on the shooting time information of the target person's face, the background content of the image, and the person's facial features, the target person's image is sequentially filtered by shooting time, shooting location, and facial features. Then, from the shooting time information of the finally filtered target person's facial image, the check-in start time and check-in end time of the target person are obtained, thereby determining whether the target person's check-in is valid. Based on the above determination result, a corresponding notification message is sent to the target person's terminal device. The above check-in method uses distributed cameras to centrally capture the target person's facial images, which makes it convenient for staff to turn on the cameras anytime and anywhere. By combining the shooting time information, the background content of the image, and the multiple filtering methods of facial features, the target person's facial images that meet the corresponding conditions are retained. This can effectively prevent impersonation check-in and improve the authenticity of check-in information.

[0098] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A check-in method based on facial recognition, characterized in that, It includes the following steps: Step S1: Read the facial images of the target person captured by each of the several cameras and their capture time information; based on the capture time information, determine whether the capture time of each facial image of the target person meets the preset time condition. Step S2: Extract the background content of the images from all the target personnel face images that meet the preset time conditions, and filter out all the target personnel face images that meet the preset location conditions based on the background content. Step S3: Extract facial features of the target personnel from all facial images that meet the preset location conditions, and filter out all facial images of the target personnel that meet the preset personnel identity conditions based on the facial features. Step S4: Based on the shooting time information of the facial images of all target personnel that meet the preset personnel identity conditions, obtain the check-in start time and check-in end time of the target personnel, and thus determine whether the check-in of the target personnel is valid; then, based on the above judgment result, send the corresponding notification message to the terminal device of the target personnel. In step S1, before reading the facial image of the target person and its shooting time information, the following steps are also included: The system instructs the target person's terminal device to connect to a camera, and uses the terminal device to control the camera to take pictures of the target person and obtain an image of the target person's face. Upload the captured facial images of the target personnel and their shooting time information to the check-in management platform; Based on the chronological order of the capture time of each target person's facial image, the attendance management platform is instructed to arrange all uploaded target person's facial images in sequence to form an image set sequence; In step S1, after instructing the target person's terminal device to connect to the camera, and using the terminal device to control the camera to capture an image of the target person's face, the process further includes: Several supplementary lights are installed on each camera, so that all the supplementary lights are evenly distributed around the camera, and the direction of the light illumination of the supplementary lights is adjusted to be the same as the shooting direction of the camera. Based on the RGB values ​​and brightness distribution of the captured facial image of the target person, corresponding supplementary lighting operations are performed on the target person, and new facial images of the target person are captured again during the above supplementary lighting operations. The process is as follows: Step S101: Using the formula (1) below, obtain the color distribution value of the target person's face image based on the RGB values ​​of the captured image. In the above formula (1), D represents the color distribution value of the target person's face image; [R(i,j),G(i,j),B(i,j)] represents the RGB value of the pixel in the i-th row and j-th column of the target person's face image; m represents the number of pixels in each row of the target person's face image; n represents the number of pixels in each column of the target person's face image. This means substituting the values ​​of i from 1 to n and j from 1 to m into the parentheses to get the maximum value inside the parentheses; || means to calculate the absolute value; Step S102: After performing pixel grayscale processing on the target person's face image, the brightness distribution value of the target person's face image is obtained using the following formula (2) based on the grayscale values ​​of the target person's face image after pixel grayscale processing. In the above formula (2), G represents the brightness distribution value of the target person's face image; H(i,j) represents the gray value of the pixel in the i-th row and j-th column after pixel grayscale processing of the target person's face image; H M This represents the maximum grayscale value, which is 255. Step S103: Using the following formula (3), based on the color distribution and brightness distribution values ​​of the target person's facial image and the total number of fill lights installed on the camera, control the number of fill lights installed on the camera to be lit. In the above formula (3), q represents the number of fill lights installed on the camera; Q represents the total number of fill lights installed on the camera; Based on the number q of fill lights installed on the camera, the fill lights are controlled to cycle through the lighting process. The process is as follows: starting from the fill light installed at the top of the camera, q fill lights are lit clockwise. Then, after a certain period of time, the (q+1)th fill light clockwise is lit and the first fill light is turned off. This process is repeated to cycle through the clockwise lighting.

2. The check-in method based on face recognition as described in claim 1, characterized in that: In step S1, determining whether the shooting time of each target person's facial image meets the preset time condition based on the shooting time information specifically includes: The capture time of each target person's facial image is obtained, and the capture time is compared with a preset work start time range. If the capture time is within the preset work start time range, the corresponding target person's facial image is determined to meet the preset time condition; otherwise, the corresponding target person's facial image is determined not to meet the preset time condition.

3. The check-in method based on face recognition as described in claim 1, characterized in that: In step S2, the background content of the image extracted from all target personnel face images that meet the preset time conditions specifically includes: Identify the contour boundary between the target person and the background environment in the image from all facial images of the target person that meet the preset time conditions; Based on the outline boundary line, the background content of the image is extracted from the facial image of the target person.

4. The check-in method based on face recognition as described in claim 3, characterized in that: In step S2, filtering out all target person facial images that meet the preset location conditions based on the background image content specifically includes: Information on the types of background objects present in the background environment of the image of the target person's face is obtained from the background image content; Based on the background object type information, determine the actual shooting location of the target person's face image, and thereby determine whether the actual shooting location belongs to the preset shooting location; If the location is specified, the corresponding target person's facial image is determined to meet the preset location conditions; otherwise, the corresponding target person's facial image is determined not to meet the preset location conditions.

5. The check-in method based on face recognition as described in claim 1, characterized in that: In step S3, extracting facial features from all target personnel facial images that meet the preset location conditions specifically includes: Facial features of individuals are extracted from facial images of all target individuals that meet the predetermined location conditions.

6. The check-in method based on face recognition as described in claim 5, characterized in that: In step S3, filtering out all target person facial images that meet the preset person identity conditions based on the person's facial features specifically includes: The facial features of the person are compared with a preset database of facial features of staff members to determine the identity information of the target person corresponding to each target person's facial image. Based on the target personnel's identity information, facial images of target personnel that were not taken by staff members are removed, while facial images of target personnel that were taken by staff members are retained, thereby filtering out all facial images of target personnel that meet the preset personnel identity conditions.

7. The check-in method based on face recognition as described in claim 1, characterized in that: In step S4, based on the shooting time information corresponding to the facial images of all target personnel who meet the preset personnel identity conditions, the start and end times of the target personnel's check-in are obtained, thereby determining whether the check-in is valid; then, based on the above determination result, a corresponding notification message is sent to the target personnel's terminal device, specifically including: From the shooting time information corresponding to the facial images of all target personnel who meet the preset personnel identity conditions, the earliest and latest shooting times of the same target personnel are selected and used as the start and end times of the target personnel's check-in. Determine whether the time difference between the check-in end time and the check-in start time is greater than or equal to a preset time range threshold. If yes, the check-in of the target person is deemed valid; otherwise, the check-in of the target person is deemed invalid. If the target person's clock-in is invalid, a clock-in invalidation notification message will be sent to the target person's terminal device.

Images