Doze alarm method and system based on face detection

Abstract

The invention discloses a doze alarm method and system based on face detection, and relates to the field of doze alarm, and the system comprises a working data acquisition module, a data detection module, a preliminary analysis module, a tracking monitoring module and a comprehensive analysis module. Scientific data basis is provided for doze alarm analysis, a security check boot operator is reminded at any time, and the doze state of the target worker is effectively analyzed by analyzing the face data of the target worker and systematically analyzing the blink data and yawn data of the target worker, so that the security check boot operator can be reminded at any time. The anti-sleepiness alarm system has the advantages that the anti-sleepiness alarm system is provided, reference basis is provided for follow-up analysis of the target worker sleepiness alarm, the accuracy and effectiveness of the target worker sleepiness alarm are effectively provided, the anti-sleepiness alarm system is used for reminding and warning, and working errors can be avoided in time while the working state of the anti-sleepiness alarm system is guaranteed.

Description

Technical Field

[0001] The present application relates to the field of drowsiness alarm, and in particular to a drowsiness alarm method and system based on face detection. Background Art

[0002] In order to ensure the safety of passengers, improve the quality of security inspection services, increase passenger satisfaction, and promote the healthy development of civil aviation, the Security Inspection Department focuses on the goal of improving the quality and efficiency of civil aviation security inspections, and accelerates the application of high-tech and equipment in the field of civil aviation security inspections. The anti-drowsiness mechanism for security inspection operators still has the following deficiencies; Existing technologies have previously provided indirect management of duty status, which can only be supervised and inspected through employee safety education and training, disciplinary constraints, irregular inspections by security quality control officers and team leaders, and video surveillance reviews. However, the supervision effect is usually delayed and problems are prone to recurring, further increasing the safety risks of airports. Summary of the invention

[0003] The object of the present invention is to provide a drowsiness alarm method and system based on face detection to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a drowsiness alarm system based on face detection, comprising: Work data acquisition module: used to acquire the work log of the staff and obtain the work data of the target staff; Data detection module: used to perform image detection on the target staff and obtain the detection data of the target staff; Preliminary analysis module: used to combine and analyze the detection data and work data of the target staff to obtain the preliminary sleepiness alarm coefficient corresponding to the target staff; Tracking and monitoring module: used to perform a reaction test on the target staff according to the initial sleepiness alarm coefficient corresponding to the target staff, and obtain the corresponding reaction coefficient of the target staff; Comprehensive analysis module: used to conduct comprehensive analysis based on the preliminary drowsiness alarm coefficient corresponding to the target staff, the reaction coefficient corresponding to the target staff and the detection data of the target staff, so as to obtain the drowsiness alarm result corresponding to the target staff.

[0005] In a preferred embodiment of this scheme, the specific implementation method of the working data acquisition module is as follows: Obtain the work log of the target staff member, filter and obtain the target staff member's working hours and the number of events processed according to the target staff member's work log and the detection time point, and record the target staff member's working hours and the number of events processed as the target staff member's work data.

[0006] In a preferred embodiment of this scheme, the specific implementation method of the data detection module is as follows: The target staff member's face is continuously imaged by a visual camera preset in front of the target staff member to obtain continuous facial images corresponding to the target staff member, the facial images of the target staff member are divided into regions to obtain eye sub-images and mouth sub-images corresponding to the target staff member, and continuous eye sub-images and continuous mouth sub-images corresponding to the target staff member are obtained by statistics; The continuous eye sub-images, continuous eyebrow sub-images and continuous mouth sub-images corresponding to the target staff member are recorded as the detection data corresponding to the target staff member.

[0007] In the preferred embodiment of this scheme, the specific implementation method of the preliminary analysis module is as follows: Establish a data extraction relationship between the preliminary analysis module and the database, extract the work fatigue index versus working time curve stored in the database, and obtain the work fatigue index corresponding to the target staff member according to the working hours of the target staff member; By calculating the formula , calculate the sleepiness alarm impact coefficient of the target staff ,in It is expressed as the work fatigue index corresponding to the target staff, Expressed as the number of processed events by the target worker, It is the working hours of the target staff. It is expressed as a preset sleepiness alarm influence factor; Extracting the eye-closing action model and the yawning action model of the character stored in the database; Perform eye closing recognition on the continuous eye sub-images corresponding to the target staff member according to the eye closing action model of the person stored in the database, obtain each eye closing corresponding to the target staff member, perform image extraction on each eye closing corresponding to the target staff member, and obtain the eye sub-image of the start of eye closing, the eye sub-image of the end of eye closing and the eye sub-image of the open eye corresponding to each eye closing of the target staff member; Acquire image data of the eye sub-image at the beginning of eye closure, the eye sub-image at the end of eye closure, and the eye sub-image at the opening of the eyes, and obtain the length and width of the eye fissure at the beginning of eye closure, the length of the eye fissure at the end of eye closure, and the length and width of the eye fissure at the opening of the eyes; By calculating the formula , calculate the eye closure depth coefficient corresponding to each eye closure ; The difference in length and width of the palpebral fissure at the beginning of eye closure and at the beginning of eye opening corresponding to each eye closure is obtained by calculation; By calculating the formula , calculate the eye opening amplitude coefficient corresponding to each eye closing ,in , They are respectively represented by the length of the palpebral fissure at the beginning of eye closure and the length of the palpebral fissure at the end of eye closure corresponding to each eye closure. It is expressed as the width of the palpebral fissure at the beginning of each eye closure. , They are respectively expressed as the difference in length and width of the palpebral fissure at the beginning of eye closure and at the beginning of eye opening. It is represented by the number of each eye closure; By calculating the formula , calculate the initial warning coefficient corresponding to each eye closure ; Obtain the image acquisition time point of the eye sub-image corresponding to the start of eye closing and the image acquisition time point of the eye sub-image corresponding to the opening of the eyes during each eye closing, and calculate the eye closing time corresponding to each eye closing; Obtain the duration of the alarm judgment time period corresponding to the alarm system, and divide it according to the duration of the alarm judgment time period corresponding to the alarm system and the image acquisition time point of the eye sub-image corresponding to the start of eye closing in each eye closing, to obtain the number of eye closing times corresponding to the alarm judgment time period; Obtain a data comparison set corresponding to the alarm coefficient, wherein the data comparison set includes a standard eye closing frequency, By calculating the formula , calculate the initial drowsiness alarm coefficient corresponding to the target staff ,in Expressed as the standard eye closing frequency, It indicates the number of eye closures corresponding to the alarm judgment time period. Indicates the duration of the alarm judgment period. It is expressed as the initial alarm coefficient corresponding to each eye closure during the alarm judgment period. It indicates the duration of eye closure corresponding to each eye closure during the alarm judgment period. It is expressed as the preset standard time spent closing eyes. Indicates the number of each eye closure during the alarm judgment period.

[0008] In the preferred embodiment of this scheme, the specific implementation method of the tracking and monitoring module is as follows: Establishing a data extraction relationship between the tracking and monitoring module and the database, extracting the fatigue inspection data set corresponding to each preliminary sleepiness alarm coefficient stored in the database, wherein the fatigue inspection data set includes the inspection character length, the inspection character brightness, the inspection character duration and the inspection character movement model; The fatigue inspection data set corresponding to the target staff is obtained by screening the preliminary sleepiness alarm coefficient corresponding to the target staff; Perform fatigue inspection through the display terminal corresponding to the target worker and the fatigue inspection data set corresponding to the target worker, and obtain the display coordinate point and display time point of the initial inspection character corresponding to the display terminal; The eye movement of the target staff member is detected by a preset visual camera in front of the target staff member to obtain the eye movement data of the target staff member, wherein the eye movement data includes the eye movement direction and eye movement speed at each time point, and the eye movement data of the target staff member after viewing the character display time point is obtained by screening, and recorded as the eye movement data to be analyzed corresponding to the target staff member; A computer vision model is established according to the to-be-analyzed eye movement data corresponding to the target staff member and the work environment corresponding to the target staff member, and a visual target point, a position of the visual target point, and a moving speed of the visual target point of the display terminal corresponding to the target staff member are obtained through the computer vision model; Extract the theoretical reaction time corresponding to each character length and character brightness combination stored in the database, and obtain the theoretical reaction time corresponding to the target staff member according to the fatigue inspection data set corresponding to the target staff member; Perform visual coincidence matching on the visual target point of the display terminal corresponding to the target staff member and the display coordinate point of the initial inspection character, obtain the time point at which the visual target point coincides with the display coordinate point of the initial inspection character, record it as the visual coincidence point corresponding to the target staff member, and record the time interval between the display time point of the initial inspection character and the visual coincidence point corresponding to the target staff member as the reaction time corresponding to the target staff member; When the system recognizes that the visual target point coincides with the display coordinate point of the initial inspection character, the inspection character moves according to the inspection character movement model and movement speed, obtains the time point when the inspection character reaches the inspection character end point, obtains the time point when the visual target point reaches the inspection character end point, and records the time interval between the time point when the inspection character reaches the inspection character end point and the time point when the visual target point reaches the inspection character end point as the delay time corresponding to the target staff member; A visual movement model corresponding to the target staff member is established according to the position of the visual target point and the moving speed of the visual target point, and a model coincidence analysis is performed on the visual movement model corresponding to the target staff member and the inspection character movement model corresponding to the target staff member to obtain the coincidence degree of the visual movement model corresponding to the target staff member; According to the calculation formula , calculate the corresponding reaction coefficient of the target staff ,in It is expressed as the theoretical response time of the target staff. It is represented by the corresponding reaction time of the target staff. It is represented by the delay time corresponding to the target worker. It is expressed as the overlap of the visual movement model corresponding to the target worker.

[0009] In the preferred embodiment of this scheme, the specific implementation method of the comprehensive analysis module is as follows: Performing action matching with the continuous mouth sub-images corresponding to the target staff member according to the yawn action model of the person stored in the database, obtaining each yawn of the continuous mouth sub-images corresponding to the target staff member, and obtaining the duration and mouth corner angle corresponding to each yawn; Filter the yawns corresponding to the alarm judgment time period and the duration and mouth angle of each yawn; By calculating the formula , calculate the comprehensive sleepiness alarm coefficient corresponding to the target staff ,in It is expressed as the number of yawns during the alarm judgment period, , They represent the duration and mouth angle of each yawn during the alarm judgment period. Indicates the number of each yawn during the alarm judgment period; The comprehensive sleepiness alarm coefficient corresponding to the target staff member is compared and analyzed with the preset standard comprehensive sleepiness alarm coefficient. If the comprehensive sleepiness alarm coefficient corresponding to the target staff member is greater than the preset standard comprehensive sleepiness alarm coefficient, the target staff member is alarmed through the alarm device. If the comprehensive sleepiness alarm coefficient corresponding to the target staff member is less than or equal to the preset standard comprehensive sleepiness alarm coefficient, the target staff member is not alarmed.

[0010] To achieve the above object, the present invention also provides the following technical solution: a drowsiness alarm method based on face detection, comprising the following steps: Obtain the work log of the staff and obtain the work data of the target staff; Perform image detection on the target staff to obtain detection data of the target staff; According to the combined analysis of the detection data and the work data of the target staff, the preliminary sleepiness alarm coefficient corresponding to the target staff is obtained; Conducting a reaction test on the target staff member according to the initial sleepiness alarm coefficient corresponding to the target staff member, and obtaining the corresponding reaction coefficient of the target staff member; A comprehensive analysis is performed based on the preliminary drowsiness alarm coefficient corresponding to the target staff, the reaction coefficient corresponding to the target staff and the detection data of the target staff to obtain the drowsiness alarm result corresponding to the target staff.

[0011] Compared with the prior art, the present invention has the following beneficial effects: The present invention collects data on the face of the staff through a visual camera, provides a scientific data basis for drowsiness alarm analysis, further improves the accuracy and effectiveness of the drowsiness alarm for the target staff, and is conducive to reminding the security operator to be in the security operator position at all times; This method effectively analyzes the sleepiness state of the target staff by analyzing the facial data of the target staff and systematically analyzing the blinking data and yawning data of the target staff, and provides a reference basis for the subsequent analysis of the sleepiness alarm of the target staff, and effectively provides the accuracy and effectiveness of the sleepiness alarm of the target staff. Through the reminders and warnings of the anti-drowsiness alarm system, while ensuring the work status of the staff themselves, work errors can be avoided in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0013] Figure 1 This is a schematic diagram of module connection according to an embodiment of the present invention.

[0014] Figure 2 It is a schematic diagram of the connection steps of an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] See also Figure 1 ,The present invention provides a drowsiness alarm system based on face detection, including a working data acquisition module, a data detection module, a preliminary analysis module, a tracking and monitoring module and a comprehensive analysis module; The working data acquisition module is connected with the preliminary analysis module, the data detection module is connected with the preliminary analysis module and the comprehensive analysis module, the preliminary analysis module is connected with the tracking and monitoring module and the comprehensive analysis module, and the tracking and monitoring module is connected with the comprehensive analysis module; The work data acquisition module is used to acquire the work log of the staff and obtain the work data of the target staff; Furthermore, the specific execution method of the work data acquisition module is as follows: Obtain the work log of the target staff member, filter and obtain the target staff member's working hours and the number of events processed according to the target staff member's work log and the detection time point, and record the target staff member's working hours and the number of events processed as the target staff member's work data.

[0017] The data detection module is used to perform image detection on the target staff and obtain the detection data of the target staff; Furthermore, the specific execution mode of the data detection module is as follows: The target staff member's face is continuously imaged by a visual camera preset in front of the target staff member to obtain continuous facial images corresponding to the target staff member, the facial images of the target staff member are divided into regions to obtain eye sub-images and mouth sub-images corresponding to the target staff member, and continuous eye sub-images and continuous mouth sub-images corresponding to the target staff member are obtained by statistics; The continuous eye sub-images, continuous eyebrow sub-images and continuous mouth sub-images corresponding to the target staff member are recorded as the detection data corresponding to the target staff member.

[0018] The preliminary analysis module is used to perform combined analysis based on the detection data of the target staff and the work data of the target staff to obtain the preliminary sleepiness alarm coefficient corresponding to the target staff; Furthermore, the specific execution method of the preliminary analysis module is as follows: Establish a data extraction relationship between the preliminary analysis module and the database, extract the work fatigue index versus working time curve stored in the database, and obtain the work fatigue index corresponding to the target staff member according to the working hours of the target staff member; By calculating the formula , calculate the sleepiness alarm impact coefficient of the target staff ,in It is expressed as the work fatigue index corresponding to the target staff, Expressed as the number of processed events by the target worker, It is the working hours of the target staff. It is expressed as a preset sleepiness alarm influence factor; Extracting the eye-closing action model and the yawning action model of the character stored in the database; Perform eye closing recognition on the continuous eye sub-images corresponding to the target staff member according to the eye closing action model of the person stored in the database, obtain each eye closing corresponding to the target staff member, perform image extraction on each eye closing corresponding to the target staff member, and obtain the eye sub-image of the start of eye closing, the eye sub-image of the end of eye closing and the eye sub-image of the open eye corresponding to each eye closing of the target staff member; Acquire image data of the eye sub-image at the beginning of eye closure, the eye sub-image at the end of eye closure, and the eye sub-image at the opening of the eyes, and obtain the length and width of the eye fissure at the beginning of eye closure, the length of the eye fissure at the end of eye closure, and the length and width of the eye fissure at the opening of the eyes; By calculating the formula , calculate the eye closure depth coefficient corresponding to each eye closure ; The difference in length and width of the palpebral fissure at the beginning of eye closure and the beginning of eye opening corresponding to each eye closure is obtained by calculation; By calculating the formula , calculate the eye opening amplitude coefficient corresponding to each eye closing ,in , They are respectively represented by the length of the palpebral fissure at the beginning of eye closure and the length of the palpebral fissure at the end of eye closure corresponding to each eye closure. It is expressed as the width of the palpebral fissure at the beginning of each eye closure. , They are respectively expressed as the difference in length and width of the palpebral fissure at the beginning of eye closure and at the beginning of eye opening. It is represented by the number of each eye closure; By calculating the formula , calculate the initial warning coefficient corresponding to each eye closure ; Obtain the image acquisition time point of the eye sub-image corresponding to the start of eye closing and the image acquisition time point of the eye sub-image corresponding to the opening of the eyes during each eye closing, and calculate the eye closing time corresponding to each eye closing; Obtain the duration of the alarm judgment time period corresponding to the alarm system, and divide it according to the duration of the alarm judgment time period corresponding to the alarm system and the image acquisition time point of the eye sub-image corresponding to the start of eye closing in each eye closing, to obtain the number of eye closing times corresponding to the alarm judgment time period; Obtain a data comparison set corresponding to the alarm coefficient, wherein the data comparison set includes a standard eye closing frequency, By calculating the formula , calculate the initial drowsiness alarm coefficient corresponding to the target staff ,in Expressed as the standard eye closing frequency, It indicates the number of eye closures corresponding to the alarm judgment time period. Indicates the duration of the alarm judgment period. It is expressed as the initial alarm coefficient corresponding to each eye closure during the alarm judgment period. It indicates the duration of eye closure corresponding to each eye closure during the alarm judgment period. It is expressed as the preset standard time spent closing eyes. Indicates the number of each eye closure during the alarm judgment period.

[0019] The tracking and monitoring module is used to perform a reaction test on the target staff member according to the initial sleepiness alarm coefficient corresponding to the target staff member, and obtain the reaction coefficient corresponding to the target staff member; Furthermore, the specific execution mode of the tracking and monitoring module is as follows: Establishing a data extraction relationship between the tracking and monitoring module and the database, extracting the fatigue inspection data set corresponding to each preliminary sleepiness alarm coefficient stored in the database, wherein the fatigue inspection data set includes the inspection character length, the inspection character brightness, the inspection character duration and the inspection character movement model; It should be noted that the inspection character movement model refers to the inspection character moving regularly in the display terminal, such as drawing a circle clockwise and drawing a square counterclockwise.

[0020] The fatigue inspection data set corresponding to the target staff is obtained by screening the preliminary sleepiness alarm coefficient corresponding to the target staff; Perform fatigue inspection through the display terminal corresponding to the target worker and the fatigue inspection data set corresponding to the target worker, and obtain the display coordinate point and display time point of the initial inspection character corresponding to the display terminal; The eye movement of the target staff member is detected by a preset visual camera in front of the target staff member to obtain the eye movement data of the target staff member, wherein the eye movement data includes the eye movement direction and eye movement speed at each time point, and the eye movement data of the target staff member after viewing the character display time point is obtained by screening, and recorded as the eye movement data to be analyzed corresponding to the target staff member; A computer vision model is established according to the to-be-analyzed eye movement data corresponding to the target staff member and the work environment corresponding to the target staff member, and a visual target point, a position of the visual target point, and a moving speed of the visual target point of the display terminal corresponding to the target staff member are obtained through the computer vision model; Extract the theoretical reaction time corresponding to each character length and character brightness combination stored in the database, and obtain the theoretical reaction time corresponding to the target staff member according to the fatigue inspection data set corresponding to the target staff member; Perform visual coincidence matching on the visual target point of the display terminal corresponding to the target staff member and the display coordinate point of the initial inspection character, obtain the time point at which the visual target point coincides with the display coordinate point of the initial inspection character, record it as the visual coincidence point corresponding to the target staff member, and record the time interval between the display time point of the initial inspection character and the visual coincidence point corresponding to the target staff member as the reaction time corresponding to the target staff member; When the system recognizes that the visual target point coincides with the display coordinate point of the initial inspection character, the inspection character moves according to the inspection character movement model and movement speed, obtains the time point when the inspection character reaches the inspection character end point, obtains the time point when the visual target point reaches the inspection character end point, and records the time interval between the time point when the inspection character reaches the inspection character end point and the time point when the visual target point reaches the inspection character end point as the delay time corresponding to the target staff member; A visual movement model corresponding to the target staff member is established according to the position of the visual target point and the moving speed of the visual target point, and a model coincidence analysis is performed on the visual movement model corresponding to the target staff member and the inspection character movement model corresponding to the target staff member to obtain the coincidence degree of the visual movement model corresponding to the target staff member; According to the calculation formula , calculate the corresponding reaction coefficient of the target staff ,in It is expressed as the theoretical response time of the target staff. It is represented by the corresponding reaction time of the target staff. It is represented by the delay time corresponding to the target worker. It is expressed as the overlap of the visual movement model corresponding to the target worker.

[0021] Comprehensive analysis module: used to conduct comprehensive analysis based on the preliminary drowsiness alarm coefficient corresponding to the target staff, the reaction coefficient corresponding to the target staff and the detection data of the target staff, so as to obtain the drowsiness alarm result corresponding to the target staff.

[0022] Furthermore, the specific implementation of the comprehensive analysis module is as follows: Performing action matching with the continuous mouth sub-images corresponding to the target staff member according to the yawn action model of the person stored in the database, obtaining each yawn of the continuous mouth sub-images corresponding to the target staff member, and obtaining the duration and mouth corner angle corresponding to each yawn; Filter the yawns corresponding to the alarm judgment time period and the duration and mouth angle of each yawn; By calculating the formula , calculate the comprehensive sleepiness alarm coefficient corresponding to the target staff ,in It is expressed as the number of yawns during the alarm judgment period, , They represent the duration and mouth angle of each yawn during the alarm judgment period. Indicates the number of each yawn during the alarm judgment period; The comprehensive sleepiness alarm coefficient corresponding to the target staff member is compared and analyzed with the preset standard comprehensive sleepiness alarm coefficient. If the comprehensive sleepiness alarm coefficient corresponding to the target staff member is greater than the preset standard comprehensive sleepiness alarm coefficient, the target staff member is alarmed through the alarm device. If the comprehensive sleepiness alarm coefficient corresponding to the target staff member is less than or equal to the preset standard comprehensive sleepiness alarm coefficient, the target staff member is not alarmed.

[0023] See also Figure 2 To achieve the above purpose, the present invention also provides the following technical solution: a drowsiness alarm method based on face detection, comprising the following steps: Obtain the work log of the staff and obtain the work data of the target staff; Perform image detection on the target staff to obtain detection data of the target staff; According to the combined analysis of the detection data and the work data of the target staff, the preliminary sleepiness alarm coefficient corresponding to the target staff is obtained; Conducting a reaction test on the target staff member according to the initial sleepiness alarm coefficient corresponding to the target staff member, and obtaining the corresponding reaction coefficient of the target staff member; A comprehensive analysis is performed based on the preliminary drowsiness alarm coefficient corresponding to the target staff, the reaction coefficient corresponding to the target staff and the detection data of the target staff to obtain the drowsiness alarm result corresponding to the target staff.

[0024] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A drowsiness alarm system based on face detection, characterized in that: include: Work data acquisition module: used to acquire the work log of the staff and obtain the work data of the target staff; Data detection module: used to perform image detection on the target staff and obtain the detection data of the target staff; Preliminary analysis module: used to combine and analyze the detection data and work data of the target staff to obtain the preliminary sleepiness alarm coefficient corresponding to the target staff; Tracking and monitoring module: used to perform a reaction test on the target staff according to the initial sleepiness alarm coefficient corresponding to the target staff, and obtain the corresponding reaction coefficient of the target staff; Comprehensive analysis module: used to conduct comprehensive analysis based on the preliminary drowsiness alarm coefficient corresponding to the target staff, the reaction coefficient corresponding to the target staff and the detection data of the target staff, so as to obtain the drowsiness alarm result corresponding to the target staff.

2. The face detection-based drowsiness alarm system according to claim 1, characterized in that: The specific implementation method of the working data acquisition module is as follows: Obtain the work log of the target staff member, filter and obtain the target staff member's working hours and the number of events processed according to the target staff member's work log and the detection time point, and record the target staff member's working hours and the number of events processed as the target staff member's work data.

3. The face detection-based drowsiness alarm system according to claim 2, characterized in that: The specific implementation method of the data detection module is as follows: The target staff member's face is continuously imaged by a visual camera preset in front of the target staff member to obtain continuous facial images corresponding to the target staff member, the facial images of the target staff member are divided into regions to obtain eye sub-images and mouth sub-images corresponding to the target staff member, and continuous eye sub-images and continuous mouth sub-images corresponding to the target staff member are obtained by statistics; The continuous eye sub-images and continuous mouth sub-images corresponding to the target staff member are recorded as the detection data corresponding to the target staff member.

4. The face detection-based drowsiness alarm system according to claim 3, characterized in that: The specific implementation method of the preliminary analysis module is as follows: Establish a data extraction relationship between the preliminary analysis module and the database, extract the work fatigue index versus working time curve stored in the database, and obtain the work fatigue index corresponding to the target staff member according to the working hours of the target staff member; By calculating the formula , calculate the sleepiness alarm impact coefficient of the target staff ,in It is expressed as the work fatigue index corresponding to the target staff, Expressed as the number of processed events by the target worker, It is the working hours of the target staff. It is expressed as a preset sleepiness alarm influence factor; Extracting the eye-closing action model and the yawning action model of the character stored in the database; Perform eye closing recognition on the continuous eye sub-images corresponding to the target staff member according to the eye closing action model of the person stored in the database, obtain each eye closing corresponding to the target staff member, perform image extraction on each eye closing corresponding to the target staff member, and obtain the eye sub-image of the start of eye closing, the eye sub-image of the end of eye closing and the eye sub-image of the open eye corresponding to each eye closing of the target staff member; Acquire image data of the eye sub-image at the beginning of eye closure, the eye sub-image at the end of eye closure, and the eye sub-image at the opening of the eyes, and obtain the length and width of the eye fissure at the beginning of eye closure, the length of the eye fissure at the end of eye closure, and the length and width of the eye fissure at the opening of the eyes; By calculating the formula , calculate the eye closure depth coefficient corresponding to each eye closure ; The difference in length and width of the palpebral fissure at the beginning of eye closure and at the beginning of eye opening corresponding to each eye closure is obtained by calculation; By calculating the formula , calculate the eye opening amplitude coefficient corresponding to each eye closing ,in , They are respectively represented by the length of the palpebral fissure at the beginning of eye closure and the length of the palpebral fissure at the end of eye closure corresponding to each eye closure. It is expressed as the width of the palpebral fissure at the beginning of each eye closure. , They are respectively expressed as the difference in length and width of the palpebral fissure at the beginning of eye closure and at the beginning of eye opening. It is represented by the number of each eye closure; By calculating the formula , calculate the initial warning coefficient corresponding to each eye closure ; Obtain the image acquisition time point of the eye sub-image corresponding to the start of eye closing and the image acquisition time point of the eye sub-image corresponding to the opening of the eyes during each eye closing, and calculate the eye closing time corresponding to each eye closing; Obtain the duration of the alarm judgment time period corresponding to the alarm system, and divide it according to the duration of the alarm judgment time period corresponding to the alarm system and the image acquisition time point of the eye sub-image corresponding to the start of eye closing in each eye closing, to obtain the number of eye closing times corresponding to the alarm judgment time period; Obtain a data comparison set corresponding to the alarm coefficient, wherein the data comparison set includes a standard eye closing frequency, By calculating the formula , calculate the initial drowsiness alarm coefficient corresponding to the target staff ,in Expressed as the standard eye-closing frequency, It indicates the number of eye closures corresponding to the alarm judgment time period. Indicates the duration of the alarm judgment period. It is expressed as the initial alarm coefficient corresponding to each eye closure during the alarm judgment period. It indicates the duration of eye closure corresponding to each eye closure during the alarm judgment period. It is expressed as the preset standard time spent closing eyes. Indicates the number of each eye closure during the alarm judgment period.

5. The face detection-based drowsiness alarm system according to claim 4, characterized in that: The specific implementation of the tracking and monitoring module is as follows: Establishing a data extraction relationship between the tracking and monitoring module and the database, extracting the fatigue inspection data set corresponding to each preliminary sleepiness alarm coefficient stored in the database, wherein the fatigue inspection data set includes the inspection character length, the inspection character brightness, the inspection character duration and the inspection character movement model; The fatigue inspection data set corresponding to the target staff is obtained by screening the preliminary sleepiness alarm coefficient corresponding to the target staff; Perform fatigue inspection through the display terminal corresponding to the target worker and the fatigue inspection data set corresponding to the target worker, and obtain the display coordinate point and display time point of the initial inspection character corresponding to the display terminal; The eye movement of the target staff member is detected by a preset visual camera in front of the target staff member to obtain the eye movement data of the target staff member, wherein the eye movement data includes the eye movement direction and eye movement speed at each time point, and the eye movement data of the target staff member after viewing the character display time point is obtained by screening, and recorded as the eye movement data to be analyzed corresponding to the target staff member; A computer vision model is established according to the to-be-analyzed eye movement data corresponding to the target staff member and the work environment corresponding to the target staff member, and a visual target point, a position of the visual target point, and a moving speed of the visual target point of the display terminal corresponding to the target staff member are obtained through the computer vision model; Extract the theoretical reaction time corresponding to each character length and character brightness combination stored in the database, and obtain the theoretical reaction time corresponding to the target staff member according to the fatigue inspection data set corresponding to the target staff member; Perform visual coincidence matching on the visual target point of the display terminal corresponding to the target staff member and the display coordinate point of the initial inspection character, obtain the time point at which the visual target point coincides with the display coordinate point of the initial inspection character, record it as the visual coincidence point corresponding to the target staff member, and record the time interval between the display time point of the initial inspection character and the visual coincidence point corresponding to the target staff member as the reaction time corresponding to the target staff member; When the system recognizes that the visual target point coincides with the display coordinate point of the initial inspection character, the inspection character moves according to the inspection character movement model and movement speed, obtains the time point when the inspection character reaches the inspection character end point, obtains the time point when the visual target point reaches the inspection character end point, and records the time interval between the time point when the inspection character reaches the inspection character end point and the time point when the visual target point reaches the inspection character end point as the delay time corresponding to the target staff member; A visual movement model corresponding to the target staff member is established according to the position of the visual target point and the moving speed of the visual target point, and a model coincidence analysis is performed on the visual movement model corresponding to the target staff member and the inspection character movement model corresponding to the target staff member to obtain the coincidence degree of the visual movement model corresponding to the target staff member; According to the calculation formula , calculate the corresponding reaction coefficient of the target staff ,in It is expressed as the theoretical response time of the target staff. It is represented by the corresponding reaction time of the target staff. It is represented by the delay time corresponding to the target worker. It is expressed as the overlap of the visual movement model corresponding to the target worker.

6. The face detection-based drowsiness alarm system according to claim 5, characterized in that: The specific implementation method of the comprehensive analysis module is as follows: Performing action matching with the continuous mouth sub-images corresponding to the target staff member according to the yawn action model of the person stored in the database, obtaining each yawn of the continuous mouth sub-images corresponding to the target staff member, and obtaining the duration and mouth corner angle corresponding to each yawn; Filter the yawns corresponding to the alarm judgment time period and the duration and mouth angle of each yawn; By calculating the formula , calculate the comprehensive sleepiness alarm coefficient corresponding to the target staff ,in It is expressed as the number of yawns during the alarm judgment period, , They represent the duration and mouth angle of each yawn during the alarm judgment period. Indicates the number of each yawn during the alarm judgment period; The comprehensive sleepiness alarm coefficient corresponding to the target staff member is compared and analyzed with the preset standard comprehensive sleepiness alarm coefficient. If the comprehensive sleepiness alarm coefficient corresponding to the target staff member is greater than the preset standard comprehensive sleepiness alarm coefficient, the target staff member is alarmed through the alarm device. If the comprehensive sleepiness alarm coefficient corresponding to the target staff member is less than or equal to the preset standard comprehensive sleepiness alarm coefficient, the target staff member is not alarmed.

7. A drowsiness alarm method based on face detection, applied to a drowsiness alarm system based on face detection as claimed in any one of claims 1 to 6, characterized in that: include: Obtain the work log of the staff and obtain the work data of the target staff; Perform image detection on the target staff to obtain detection data of the target staff; According to the combined analysis of the detection data and the work data of the target staff, the preliminary sleepiness alarm coefficient corresponding to the target staff is obtained; Conducting a reaction test on the target staff member according to the initial sleepiness alarm coefficient corresponding to the target staff member, and obtaining the corresponding reaction coefficient of the target staff member; A comprehensive analysis is performed based on the preliminary drowsiness alarm coefficient corresponding to the target staff, the reaction coefficient corresponding to the target staff and the detection data of the target staff to obtain the drowsiness alarm result corresponding to the target staff.

Images