A warning system and method for avoiding fatigue work

Abstract

The present invention provides an early warning system and method for avoiding fatigue work, relating to the field of computer technology. The system comprises a data collection module for collecting early warning data from workers; an eye state data accuracy assessment module for analyzing the accuracy of two sets of eye state assessment data based on an eye state assessment data accuracy index; a physiological state data accuracy assessment module for analyzing the accuracy of two sets of physiological state assessment data based on an physiological state assessment data accuracy index; an early warning data accuracy assessment module for obtaining an early warning data accuracy assessment index for the worker; and an analysis module for analyzing the accuracy of the early warning data based on the early warning data accuracy assessment index and issuing an early warning. This system accurately warns workers of fatigue and addresses the existing difficulty in obtaining accurate and timely early warning data from workers.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an early warning system and method for avoiding fatigue work. Background Art

[0002] In modern society, many positions require workers to concentrate their attention for long periods of time, and even involve working across the day and night. Especially when flying, driving, or operating complex equipment, long periods of work not only consume mental energy, but are also accompanied by considerable physical exertion, which will reduce the workers' alertness and reaction speed, causing a decline in work ability. High-intensity, long-term, high-cognitive load work, especially repeated work at night or across the day and night, will have a significant adverse effect on the workers' physiological rhythms and attention, and easily lead to fatigue. In severe cases, it endangers work safety and even causes accidents, which will bring irreparable losses and harm to society and individuals. Therefore, early warning systems and methods for avoiding fatigue work are of great significance to both society and individuals.

[0003] Existing fatigue work warning systems and methods use infrared cameras to capture images of workers working, then use the MTCNNN face detection method to detect facial images. These facial images are then segmented using differential image adaptive threshold segmentation to obtain eye images. Feature extraction and tracking of the eye images are then performed using a combination of PCA principal component analysis and Kalman filter eye tracking. Finally, the system uses feature extraction and tracking of the eye images to calculate the ratio of the number of eye closure frames to the total number of recognized eye image frames, thereby determining and warning the worker's fatigue status. However, using data on the worker's eye status for these warnings presents a challenge in obtaining accurate and timely data on the worker's fatigue status. Summary of the Invention

[0004] The present invention provides an early warning system and method for avoiding fatigue work, which solves the problem in the prior art that it is difficult to accurately and timely obtain effective early warning data of workers, and realizes accurate early warning of the fatigue status of workers.

[0005] In order to solve the above-mentioned purpose of the invention, the technical solution provided by the present invention is as follows:

[0006] On the one hand, an embodiment of the present invention provides an early warning system for avoiding fatigue work, comprising: a data collection module, an eye state data accuracy assessment module, a physiological state data accuracy assessment module, an early warning data accuracy assessment module and an analysis module; the data collection module collects early warning data obtained by two sets of identical equipment at two preset positions at the same time by the operator; the eye state data accuracy assessment module obtains two sets of eye state assessment data and image data in the early warning data and analyzes them to obtain eye state assessment data accuracy indicators and image quality analysis indicators, analyzes the accuracy of the two sets of eye state assessment data according to the eye state assessment data accuracy indicators, and selects the image quality analysis indicators for the early warning system according to the image quality analysis indicators. Accurate data for eye state assessment; the physiological state data accuracy assessment module: obtains two groups of physiological state assessment data in the warning data and analyzes them to obtain the physiological state assessment data accuracy index and the physiological state assessment data selection index, analyzes the accuracy of the two groups of physiological state assessment data according to the physiological state assessment data accuracy index, and determines the valid data for physiological state assessment according to the physiological state assessment data selection index; the warning data accuracy assessment module: comprehensively integrates the eye state assessment data accuracy index and the physiological state assessment data accuracy index to obtain the warning data accuracy assessment index of the operator; the analysis module: analyzes the accuracy of the warning data according to the warning data accuracy assessment index and issues a warning.

[0007] Optionally, the early warning data includes two groups of eye state assessment data and two groups of physiological state assessment data; the two groups of eye state assessment data include: a first group of eye state assessment data and a second group of eye state assessment data, the first group of eye state assessment data includes a first group of blinking frequency, a first group of blinking duration, a first group of eye movement speed, a first group of pupil diameter and a first group of eyelid closure duration data, and the second group of eye state assessment data includes a second group of blinking frequency, a second group of blinking duration, a second group of eye movement speed, a second group of pupil diameter and a second group of eyelid closure duration data; the two groups of physiological state assessment data include a first group of physiological state assessment data and a second group of physiological state assessment data, the first group of physiological state assessment data includes a first group of heart rate, a first group of skin conductivity, a first group of brain wave alpha wave number and a first group of blood oxygen saturation, and the second group of physiological state assessment data includes a second group of heart rate, a second group of skin conductivity, a second group of brain wave alpha wave number and a second group of blood oxygen saturation.

[0008] Optionally, the specific method for obtaining the accuracy index of the eye condition assessment data is:

[0009]

[0010] Wherein, α is the accuracy index of eye status assessment data, f″ is the second group blink frequency, t″ is the second group blink duration, s″ is the second group eye movement velocity, d″ is the second group pupil diameter, T″ is the second group eyelid closure duration, f′ is the first group blink frequency, t′ is the first group blink duration, s′ is the first group eye movement velocity, d′ is the first group pupil diameter, T′ is the first group eyelid closure duration, and e is a natural constant.

[0011] Optionally, the specific analysis process of the eye state assessment data accuracy index is: obtaining a preset eye state assessment data accuracy index threshold, obtaining the eye state assessment data accuracy index, comparing the obtained eye state assessment data accuracy index with the eye state assessment data accuracy index threshold; when the eye state assessment data accuracy index is greater than the eye state assessment data accuracy index threshold, the eye state assessment data accuracy index is in an abnormal range; when the eye state assessment data accuracy index is less than or equal to the eye state assessment data accuracy index threshold, the eye state assessment data accuracy index is in a normal range.

[0012] Optionally, the specific process of analyzing the accuracy of the two sets of eye state assessment data based on the eye state assessment data accuracy index is as follows: when the eye state assessment data accuracy index is within the normal range, it means that the two sets of eye state assessment data obtained are accurate, and the two sets of eye state assessment data at this time can be used to assess the eye state and issue an early warning; when the eye state assessment data accuracy index is within the abnormal range, it means that the two sets of eye state assessment data obtained at this time are inaccurate; when the eye state assessment data accuracy index is within the abnormal range, the eye state image data is obtained for analysis to obtain the image quality analysis index, and the valid data for eye state assessment is determined based on the image quality analysis index.

[0013] Optionally, the specific analysis process of the physiological state assessment data accuracy index is: obtaining a preset physiological state assessment data accuracy index threshold, obtaining the physiological state assessment data accuracy index, comparing the obtained physiological state assessment data accuracy index with the physiological state assessment data accuracy index threshold; when the physiological state assessment data accuracy index is greater than the physiological state assessment data accuracy index threshold, the physiological state assessment data accuracy index is in an abnormal range; when the physiological state assessment data accuracy index is less than or equal to the physiological state assessment data accuracy index threshold, the physiological state assessment data accuracy index is in a normal range.

[0014] Optionally, the specific process of analyzing the accuracy of the two sets of physiological state assessment data based on the physiological state assessment data accuracy index is: when the physiological state assessment data accuracy index is in the normal range, it means that the two sets of physiological state assessment data obtained are accurate; when the physiological state assessment data accuracy index is in the abnormal range, it means that the two sets of physiological state assessment data obtained are inaccurate; when the physiological state assessment data accuracy index is in the abnormal range, the physiological state assessment data selection index is obtained, and the valid data for physiological state assessment is determined according to the physiological state assessment data selection index.

[0015] Optionally, the specific analysis process of the warning data accuracy assessment indicator is: obtaining a preset warning data accuracy assessment indicator threshold, obtaining the warning data accuracy assessment indicator, comparing the obtained warning data accuracy assessment indicator with the warning data accuracy assessment indicator threshold; when the warning data accuracy assessment indicator is greater than the warning data accuracy assessment indicator threshold, the warning data accuracy assessment indicator is in an abnormal range; when the warning data accuracy assessment indicator is less than or equal to the warning data accuracy assessment indicator threshold, the warning data accuracy assessment indicator is in a normal range.

[0016] Optionally, the specific process of analyzing the accuracy of the warning data and issuing a warning based on the warning data accuracy evaluation index is: when the warning data accuracy evaluation index is within the normal range, it means that the obtained warning data is accurate, and the obtained warning data can be used for warning; when the warning data accuracy evaluation index is within the abnormal range, it means that the obtained warning data is inaccurate, and the warning data will not be used to warn the fatigue status of the operator.

[0017] On the other hand, an embodiment of the present invention also provides an early warning method for avoiding fatigue work, comprising the following steps: collecting early warning data obtained by two sets of identical equipment at two preset positions at the same time by the operator; obtaining two sets of eye state assessment data and image data in the early warning data and analyzing them to obtain an eye state assessment data accuracy index and an image quality analysis index, analyzing the accuracy of the two sets of eye state assessment data according to the eye state assessment data accuracy index, and selecting accurate data for eye state assessment according to the image quality analysis index; obtaining two sets of physiological state assessment data in the early warning data and analyzing them to obtain a physiological state assessment data accuracy index and a physiological state assessment data selection index, analyzing the accuracy of the two sets of physiological state assessment data according to the physiological state assessment data accuracy index, and determining the valid data for physiological state assessment according to the physiological state assessment data selection index; comprehensively obtaining the eye state assessment data accuracy index and the physiological state assessment data accuracy index to obtain the operator's early warning data accuracy assessment index; analyzing the accuracy of the warning data according to the early warning data accuracy assessment index and issuing an early warning.

[0018] Compared with the prior art, the above technical solution has at least the following beneficial effects:

[0019] 1. By obtaining two sets of eye state assessment data of the operator, the accuracy index of the eye state assessment data of the operator is obtained and the two sets of eye state assessment data of the operator are evaluated, and the two sets of eye state image data are obtained and analyzed to obtain the image quality analysis index. According to the image quality analysis index, effective data is selected to obtain accurate data in a timely and effective manner, and then two sets of physiological state assessment data of the operator are obtained and the accurate data for eye state assessment is determined according to the physiological state assessment data selection index, thereby obtaining the accuracy index of the physiological state assessment data of the operator and analyzing the two sets of physiological state assessment data of the operator. Finally, the accuracy index of the eye state assessment data and the accuracy index of the physiological state assessment data are comprehensively combined to obtain the accuracy assessment index of the early warning data, and the accuracy of the early warning data is accurately analyzed according to the early warning data accuracy assessment index, and then an early warning is issued based on the accurate early warning data, which effectively solves the problem in the existing technology that it is difficult to accurately and timely obtain effective early warning data for the operator.

[0020] 2. By obtaining two sets of eye status assessment data of the operating personnel, the accuracy index of the eye status assessment data of the operating personnel is obtained and the accuracy of the two sets of eye status assessment data of the operating personnel is analyzed, thereby accurately analyzing the accuracy of the two sets of eye status assessment data.

[0021] 3. By obtaining two sets of physiological status assessment data of the operating personnel, the accuracy index of the physiological status assessment data is obtained and the accuracy of the two sets of physiological status assessment data of the operating personnel is analyzed, thereby accurately analyzing the accuracy of the two sets of physiological status assessment data of the operating personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A diagram of an early warning system for avoiding fatigue work provided by the present invention;

[0024] Figure 2 A graph showing the accuracy evaluation index of early warning data provided by the present invention;

[0025] Figure 3 A graph showing comprehensive evaluation indicators of fatigue status of operators provided by the present invention;

[0026] Figure 4 This is a flow chart of an early warning method for avoiding fatigue work provided by the present invention. DETAILED DESCRIPTION

[0027] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described 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.

[0028] Unless otherwise defined, the technical or scientific terms used in the present invention shall have the usual meaning understood by persons of ordinary skill in the field to which the present invention belongs. The words "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one", "an" or "the" do not indicate a quantity limitation, but rather indicate the presence of at least one. Words such as "include" or "comprise" mean that the elements or objects preceding the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0029] It should be noted that the terms "up", "down", "left", "right", "front" and "back" used in the present invention are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0030] In order to solve the problem that it is difficult to obtain effective early warning data of operators accurately and timely, the present invention provides a system and method that can accurately analyze the fatigue status of operators and issue early warnings.

[0031] like Figure 1As shown in FIG, a diagram of an early warning system for avoiding fatigue work provided by an embodiment of the present invention includes: a data collection module, an eye state data accuracy evaluation module, a physiological state data accuracy evaluation module, an early warning data accuracy evaluation module and an analysis module; the data collection module collects early warning data obtained by two sets of identical equipment at two preset positions at the same time by the operator; the eye state data accuracy evaluation module obtains two sets of eye state evaluation data and image data in the early warning data and analyzes them to obtain eye state evaluation data accuracy indicators and image quality analysis indicators, analyzes the accuracy of the two sets of eye state evaluation data according to the eye state evaluation data accuracy indicators, and selects the image quality analysis indicator according to the image quality analysis indicator. Accurate data for eye status assessment; physiological status data accuracy assessment module: obtain two groups of physiological status assessment data in the warning data and analyze them to obtain the physiological status assessment data accuracy index and the physiological status assessment data selection index, analyze the accuracy of the two groups of eye status assessment data according to the physiological status assessment data accuracy index, and determine the accurate data for eye status assessment according to the physiological status assessment data selection index; warning data accuracy assessment module: comprehensively integrate the eye status assessment data accuracy index and the physiological status assessment data accuracy index to obtain the warning data accuracy assessment index of the operator; analysis module: analyze the accuracy of the warning data according to the warning data accuracy assessment index and issue a warning.

[0032] In this embodiment, it should be noted that the two sets of eye state assessment data are two sets of eye state assessment data obtained by the same devices placed at different positions, wherein the first set of eye state assessment data is obtained by the first set of devices at a preset first position, and the second set of eye state assessment data is obtained by the second set of devices at a preset second position. The two sets of physiological state assessment data are two sets of physiological state assessment data obtained by the same devices placed at different positions, wherein the first set of physiological state assessment data is obtained by placing it at the preset position of the first group, and the second set of physiological state assessment data is obtained by placing it at the preset position of the second group. It should be noted that the preset position of the first group and the preset position of the second group are two similar sets of positions.

[0033] Among them, the warning data include two groups of eye state assessment data and two groups of physiological state assessment data. The two groups of eye state assessment data include: a first group of eye state assessment data and a second group of eye state assessment data. The first group of eye state assessment data includes a first group of blinking frequency, a first group of blinking duration, a first group of eye movement speed, a first group of pupil diameter and a first group of eyelid closure duration data. The second group of eye state assessment data includes a second group of blinking frequency, a second group of blinking duration, a second group of eye movement speed, a second group of pupil diameter and a second group of eyelid closure duration data; the two groups of physiological state assessment data include a first group of physiological state assessment data and a second group of physiological state assessment data. The first group of physiological state assessment data includes a first group of heart rate, a first group of skin conductivity, a first group of brain wave alpha wave number and a first group of blood oxygen saturation. The second group of physiological state assessment data includes a second group of heart rate, a second group of skin conductivity, a second group of brain wave alpha wave number and a second group of blood oxygen saturation.

[0034] In this embodiment, the first group of blink frequency, the first group of blink duration, the first group of eye movement speed, the first group of pupil diameter, the first group of eyelid closure duration data, the second group of blink frequency, the second group of blink duration, the second group of eye movement speed, the second group of pupil diameter and the second group of eyelid closure duration data can be obtained through an eye tracker; the first group of heart rate and the second group of heart rate can be obtained by placing an optical sensor on the steering wheel or seat, the first group of skin conductivity and the second group of skin conductivity can be obtained by installing a non-contact sensor, such as a capacitive sensor, on the steering wheel or seat, the first group of brain wave alpha wave number and the second group of brain wave alpha wave number can be obtained through wireless dry electrode technology, and the first group of blood oxygen saturation and the second group of blood oxygen saturation can be obtained by a pulse oximeter clipped to the earlobe.

[0035] Among them, the specific method for obtaining the accuracy index of eye status assessment data is as follows:

[0036]

[0037] Wherein, α is the accuracy index of eye status assessment data, f″ is the second group blink frequency, t″ is the second group blink duration, s″ is the second group eye movement velocity, d″ is the second group pupil diameter, T″ is the second group eyelid closure duration, f′ is the first group blink frequency, t′ is the first group blink duration, s′ is the first group eye movement velocity, d′ is the first group pupil diameter, T′ is the first group eyelid closure duration, and e is a natural constant.

[0038] In this embodiment, f″ represents the number of blinks per unit time in the second set of data, t″ represents the time length from the start to the end of a blink in the second set of data, s″ represents the speed of the eyeball moving in the visual field in the second set of data, d″ represents the size of the pupil diameter in the second set of data, T″ represents the length of time the eyelid is completely closed in the second set of data, f′ represents the number of blinks per unit time in the first set of data, t′ represents the time length from the start to the end of a blink in the initial state of the operator in the first set of data, s′ represents the speed of the eyeball moving in the visual field in the initial state of the operator in the first set of data, d′ represents the size of the pupil diameter in the initial state of the operator in the first set of data, and T′ is the length of time the eyelid is completely closed in the initial state of the operator in the first set of data.

[0039] In this embodiment, the accuracy of the data obtained by evaluating the first set of eye state assessment data and the second set of eye state assessment data is comprehensively evaluated. The closer the first set of eye state assessment data is to the corresponding second set of eye state assessment data, for example, the closer the first set of blinking frequency is to the second set of blinking frequency, the more accurate the two sets of eye state assessment data obtained are. The slower the first set of blinking frequency is, the longer the blinking duration and the eyelid closure time are, and the slower the first set of eye movement speed is. This embodiment uses logarithmic operations for nonlinear transformation, which smooths the influence of outliers and is conducive to improving the accuracy and effectiveness of the eye state assessment data accuracy indicators.

[0040] Among them, the specific analysis process of the eye state assessment data accuracy index is: obtain a preset eye state assessment data accuracy index threshold, obtain the eye state assessment data accuracy index, compare the obtained eye state assessment data accuracy index with the eye state assessment data accuracy index threshold, when the eye state assessment data accuracy index is greater than the eye state assessment data accuracy index threshold, the eye state assessment data accuracy index is in an abnormal range; when the eye state assessment data accuracy index is less than or equal to the eye state assessment data accuracy index threshold, the eye state assessment data accuracy index is in a normal range.

[0041] In this embodiment, the preset eye state assessment data accuracy index threshold is α1, α1=1.07. When the eye state assessment data accuracy index is less than or equal to the eye state assessment data accuracy index threshold, that is, α≤α1, it indicates that the eye state assessment data accuracy index is in the normal range, that is, the two sets of eye state assessment data obtained are relatively accurate; when in other situations, that is, α>α1, it indicates that the eye state assessment data accuracy index is in the abnormal range, that is, the two sets of eye state assessment data obtained have a large deviation, that is, there are problems with the two sets of eye state assessment data obtained, and the images in the data acquisition process are analyzed to obtain a set of eye state assessment data with relatively good image quality for early warning.

[0042] Among them, the specific process of analyzing the accuracy of the two sets of eye status assessment data according to the eye status assessment data accuracy index is: when the eye status assessment data accuracy index is in the normal range, it means that the two sets of eye status assessment data obtained are accurate, and the two sets of eye status assessment data at this time can be used to assess the eye status and issue an early warning; when the eye status assessment data accuracy index is in the abnormal range, it means that the two sets of eye status assessment data obtained at this time are inaccurate; when the eye status assessment data accuracy index is in the abnormal range, the eye status image data is obtained for analysis to obtain the image quality analysis index, and the valid data for eye status assessment is determined according to the image quality analysis index.

[0043] In this embodiment, when the accuracy index of the eye state assessment data is in the normal range, it means that the difference between the first group of eye state assessment data and the second group of eye state assessment data is small, that is, the two groups of eye state assessment data obtained at this time are more reliable. At this time, the average value of the first group of eye state assessment data and the corresponding second group of eye state assessment data can be obtained as the eye state assessment data for evaluating the eye state of the operator. For example, the average value of the first group of blinking frequencies and the second group of blinking frequencies is obtained as the blinking frequency used in the early warning. The same is true for other data used in the early warning. When the accuracy index of the eye state assessment data is in the abnormal range, it means that the difference between the first group of eye state assessment data and the second group of eye state assessment data is large, that is, the two groups of eye state assessment data obtained at this time have certain problems. The data problems obtained at this time may be caused by multiple reasons. At this time, the images in the data acquisition process are analyzed to obtain a group of eye state assessment data with relatively good image quality for early warning.

[0044] In this embodiment, as described above, when an eye tracker is used to obtain two sets of eye state assessment data, two sets of images are generated. When the accuracy index of the eye state assessment data is within an abnormal range, relevant data of the two sets of images are obtained to evaluate the quality of the two sets of images, and the set of eye state assessment data with better image quality is obtained as the eye state assessment data used in fatigue state warning. The image quality analysis index is obtained by: obtaining the preset weight ratios of image resolution, image pixel density, image contrast, and image signal-to-noise ratio in the image quality analysis index from a database;

[0045]

[0046] Wherein, ε represents the image quality analysis index, A represents the image resolution, A1 represents the standard value of image resolution, Q represents the image pixel density, Q1 represents the standard value of image pixel density, G represents the image contrast, G1 represents the standard value of image contrast, F represents the image signal-to-noise ratio, F1 represents the standard value of image signal-to-noise ratio, h1 represents the weight ratio of image resolution in the image quality analysis index, h2 represents the weight ratio of image pixel density in the image quality analysis index, h3 represents the weight ratio of image contrast in the image quality analysis index, and h4 represents the weight ratio of image signal-to-noise ratio in the image quality analysis index.

[0047] In this embodiment, image resolution, image pixel density, image contrast, and image signal-to-noise ratio can be obtained through image analysis software, and standard values ​​for image resolution, image pixel density, image contrast, and image signal-to-noise ratio can be obtained from a database. In this embodiment, h1, h2, h3, and h4 are respectively the influence weight ratios corresponding to the preset image resolution, image pixel density, image contrast, and image signal-to-noise ratio, obtained from a preset database. These values ​​are used to represent the degree of influence of image resolution, image pixel density, image contrast, and image signal-to-noise ratio on the image quality analysis indicator. When used, the influence weight ratios corresponding to image resolution, image pixel density, image contrast, and image signal-to-noise ratio can be directly obtained from the preset database.

[0048] In this embodiment, image quality analysis indicators of a first group of images generated by a first group of eye state assessment data and image quality analysis indicators of a second group of images generated by a second group of eye state assessment data are obtained. The image quality analysis indicators of the first group of images are obtained by substituting the image resolution, image pixel density, image contrast, and image signal-to-noise ratio of the first group of images into the above formula. Similarly, the image quality analysis indicators of the second group of images can be obtained. The image quality analysis indicators of the first group of images and the image quality analysis indicators of the second group of images are compared. The group of image data with a smaller image quality analysis indicator is closer to the image standard data, that is, the group of image data with a smaller image quality analysis indicator is better, that is, the group of eye state assessment data with a smaller image quality analysis indicator is selected as the eye state assessment data used in fatigue state warning.

[0049] Among them, the specific analysis process of the physiological state assessment data accuracy index is: obtaining a preset physiological state assessment data accuracy index threshold, obtaining the physiological state assessment data accuracy index, comparing the obtained physiological state assessment data accuracy index with the physiological state assessment data accuracy index threshold; when the physiological state assessment data accuracy index is greater than the physiological state assessment data accuracy index threshold, the physiological state assessment data accuracy index is in an abnormal range; when the physiological state assessment data accuracy index is less than or equal to the physiological state assessment data accuracy index threshold, the physiological state assessment data accuracy index is in a normal range.

[0050] In this embodiment, the specific method for obtaining the accuracy index of the physiological status assessment data is as follows:

[0051]

[0052] Wherein, γ represents the accuracy index of physiological status assessment data, a″ represents the second group of heart rate, c″ represents the second group of skin conductivity, y″ represents the second group of brain wave α wave number, u″ represents the second group of blood oxygen saturation, a′ represents the first group of initial heart rate, c′ represents the first group of initial skin conductivity, y′ represents the first group of initial brain wave α wave number, and u′ represents the first group of blood oxygen saturation.

[0053] In this embodiment, a″ represents the number of heart beats per unit time in the second set of data, c″ represents the electrical conductivity of the skin surface in the second set of data, y″ represents the number of α waves in the electrical activity of the cerebral cortical nerve cells in the second set of data, u″ represents the percentage of oxygen content of hemoglobin in the blood in the second set of data, a′ represents the number of heart beats per unit time of the operator in the first set of data, c′ represents the electrical conductivity of the skin surface of the operator in the first set of data, y′ represents the number of α waves in the electrical activity of the cerebral cortical nerve cells of the operator in the first set of data, and u′ represents the percentage of oxygen content of hemoglobin in the blood of the operator in the first set of data.

[0054] In this embodiment, the accuracy of the physiological state assessment data obtained by comprehensively evaluating the first set of physiological state assessment data and the second set of physiological state assessment data is evaluated. The smaller the difference between the first set of physiological state assessment data and the second set of physiological state assessment data, the more accurate the two sets of physiological state assessment data obtained are, and the more accurate the early warning of the physiological state of the operator based on the two sets of physiological state assessment data is. This embodiment uses a square root operation to perform a nonlinear transformation, so that the accuracy index of the obtained physiological state assessment data is more explanatory.

[0055] In this embodiment, the preset physiological state assessment data accuracy index threshold value obtained is γ1, γ1=1.10. When the physiological state assessment data accuracy index is less than or equal to the physiological state assessment data accuracy index threshold value, that is, γ≤γ1, it indicates that the physiological state assessment data accuracy index is in the normal range, that is, it indicates that the two sets of physiological state assessment data obtained are accurate; when in other situations, that is, γ>γ1, it indicates that the physiological state assessment data accuracy index is in the abnormal range, that is, it indicates that the two sets of physiological state assessment data obtained are inaccurate.

[0056] Among them, the specific process of analyzing the accuracy of the two groups of physiological state assessment data according to the physiological state assessment data accuracy index is: when the physiological state assessment data accuracy index is in the normal range, it means that the two groups of physiological state assessment data obtained are accurate; when the physiological state assessment data accuracy index is in the abnormal range, it means that the two groups of physiological state assessment data obtained are inaccurate; when the physiological state assessment data accuracy index is in the abnormal range, the physiological state assessment data selection index is obtained, and the valid data for physiological state assessment is determined according to the physiological state assessment data selection index.

[0057] In this embodiment, when the physiological state assessment data accuracy index is within the normal range, it indicates that the first set of physiological state assessment data and the second set of physiological state assessment data differ slightly, that is, the accuracy of the two sets of physiological state assessment data obtained at this time is higher, that is, the two sets of physiological state assessment data obtained at this time can be used to assess the physiological state of the operator. At this time, the average value of the first set of physiological state assessment data and the corresponding second set of physiological state assessment data can be taken as the physiological state assessment data for assessing the physiological state of the operator. For example, the average value of the first set of heart rates and the second set of heart rates can be obtained as the heart rate used in the early warning. The same is true for other data used in the early warning. When the physiological state assessment data accuracy index is within the abnormal range, it indicates that the first set of physiological state assessment data and the second set of physiological state assessment data differ significantly, that is, the two sets of physiological state assessment data obtained at this moment are not suitable for use in the early warning of the fatigue state of the operator, which will result in inaccurate early warning of the physiological state of the operator. Then, the physiological state assessment data selection index of the two sets of physiological state assessment data is obtained to select a set of relatively better physiological state assessment data for early warning.

[0058] In this embodiment, when the accuracy index of the physiological state assessment data is within an abnormal range, the two sets of physiological state assessment data are respectively compared with the initial physiological data of the operator to obtain a physiological state assessment data selection index. The physiological state assessment data used in the fatigue state warning is selected based on the physiological state assessment data selection index of the first set of physiological state assessment data and the physiological state assessment data selection index of the second set of physiological state assessment data. The method for obtaining the physiological state assessment data selection index is as follows:

[0059]

[0060] In this embodiment, λ represents an indicator for selecting physiological status assessment data, a′ represents the number of heart beats per unit time of the operator in the first set of data, c′ represents the electrical conductivity of the operator's skin surface in the first set of data, y′ represents the number of α waves in the electrical activity of the operator's cerebral cortex nerve cells in the first set of data, u′ represents the percentage of oxygen content in the hemoglobin in the operator's blood in the first set of data, a1 represents the operator's initial heart rate, c1 represents the operator's initial skin conductivity, y1 represents the operator's initial number of α waves in the brain wave, and u1 represents the operator's initial blood oxygen saturation.

[0061] In this embodiment, obtaining the physiological data of the operator in the initial awake state when performing the operation is to obtain the initial physiological data of the operator. The initial physiological data include initial heart rate, initial skin conductivity, initial number of brain wave alpha waves and initial blood oxygen saturation. The initial heart rate can be obtained by placing an optical sensor on the steering wheel or seat, the initial skin conductivity can be obtained by installing a non-contact sensor, such as a capacitive sensor, on the steering wheel or seat, the initial number of brain wave alpha waves can be obtained by wireless dry electrode technology, and the initial blood oxygen saturation can be obtained by a pulse oximeter clipped to the earlobe.

[0062] In this embodiment, the first group of initial heart rate, the first group of initial skin conductivity, the first group of initial number of brain wave alpha waves, and the first group of blood oxygen saturation are compared with the initial heart rate, the initial skin conductivity, the initial number of brain wave alpha waves, and the initial blood oxygen saturation according to the above formula to obtain the physiological state assessment data selection index of the first group of physiological state assessment data. The second group of initial heart rate, the second group of initial skin conductivity, the second group of initial number of brain wave alpha waves, and the second group of blood oxygen saturation are compared with the initial heart rate, the initial skin conductivity, the initial number of brain wave alpha waves, and the initial blood oxygen saturation according to the above formula to obtain the physiological state assessment data selection index of the second group of physiological state assessment data. The physiological state assessment data selection index of the first group of physiological state assessment data is compared with the physiological state assessment data selection index of the second group of physiological state assessment data. The group of data with the smaller physiological state assessment data selection index is closer to the initial physiological data. In this case, the group of data with the smaller physiological state assessment data selection index is selected as the physiological state assessment data used in fatigue state warning.

[0063] Among them, the specific analysis process of the early warning data accuracy evaluation index is: obtain the preset early warning data accuracy evaluation index threshold, obtain the early warning data accuracy evaluation index, compare the obtained early warning data accuracy evaluation index with the early warning data accuracy evaluation index threshold, when the early warning data accuracy evaluation index is greater than the early warning data accuracy evaluation index threshold, the early warning data accuracy evaluation index is in the abnormal range; when the early warning data accuracy evaluation index is less than or equal to the early warning data accuracy evaluation index threshold, the early warning data accuracy evaluation index is in the normal range.

[0064] In this embodiment, the specific method for obtaining the early warning data accuracy evaluation index is: obtaining from the database the weight ratio of the preset eye state evaluation data accuracy index and the physiological state evaluation data accuracy index in the early warning data accuracy evaluation index;

[0065]

[0066] Where η is the early warning data accuracy evaluation index, α is the eye status evaluation data accuracy index, γ is the physiological status evaluation data accuracy index, k1 is the weight ratio of the eye status evaluation data accuracy index to the early warning data accuracy evaluation index, k2 is the weight ratio of the physiological status evaluation data accuracy index to the early warning data accuracy evaluation index, and e is a natural constant.

[0067] In this embodiment, k1 and k2 are respectively the influence weight ratios corresponding to the preset eye state assessment data accuracy index and the physiological state assessment data accuracy index obtained from the preset database, which are used to represent the numerical values ​​of the degree of influence of the eye state assessment data accuracy index and the physiological state assessment data accuracy index on the warning data accuracy assessment index. When used, the influence weight ratios corresponding to the eye state assessment data accuracy index and the physiological state assessment data accuracy index can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the eye state assessment data accuracy index and the physiological state assessment data accuracy index and their corresponding weights form a mapping set. After the real-time eye state assessment data accuracy index and the physiological state assessment data accuracy index are input into the mapping set, the corresponding weight ratios corresponding to the eye state assessment data accuracy index and the physiological state assessment data accuracy index will be obtained. In this example, the value ranges of k1 and k2 are both [0, 1], and the sum of k1 and k2 is 1.

[0068] In this embodiment, the accuracy index of the eye state assessment data and the accuracy index of the physiological state assessment data are combined to perform a comprehensive assessment of the fatigue state of the operator. When the accuracy assessment index of the early warning data is in an abnormal range, it means that the obtained early warning data is inaccurate, which will have a certain impact on the early warning of the operator. The larger the accuracy index of the eye state assessment data and the accuracy index of the physiological state assessment data, the larger the early warning data accuracy assessment index, that is, the more inaccurate the early warning data of the operator. This embodiment uses inverse tangent to perform nonlinear transformation, so that the obtained early warning data accuracy assessment index is more reliable and effective.

[0069] In this embodiment, the preset threshold value of the warning data accuracy evaluation index is η1, η1=1.01. When the warning data accuracy evaluation index is less than or equal to the warning data accuracy evaluation index threshold, that is, η≤η1, it means that the warning data accuracy evaluation index is in the normal range, that is, the obtained warning data is relatively accurate; when it is in other situations, that is, η>η1, it means that the warning data accuracy evaluation index is in the abnormal range, that is, the accuracy of the obtained warning data is low.

[0070] like Figure 2As shown, it is a curve diagram of the early warning data accuracy evaluation index provided by an embodiment of the present invention, with the eye state evaluation data accuracy index as the independent variable and the early warning data accuracy evaluation index as the dependent variable. Through the increase and decrease analysis of curve 1, curve 2 and curve 3, it can be seen that when the eye state evaluation data accuracy index is larger, the early warning data accuracy evaluation index is larger; when the eye state evaluation data accuracy index is smaller, the early warning data accuracy evaluation index is smaller. The statistical table of relevant data of the early warning data accuracy evaluation index is shown in Table 1:

[0071] Table 1 Statistics of relevant data on early warning data accuracy evaluation indicators

[0072]

[0073] It can be seen from Table 1 that for curve 1, curve 2 and curve 3, when the eye state assessment data accuracy index of curve 1, curve 2 and curve 3 is 1.00, when the physiological state assessment data accuracy index of curve 1, curve 2 and curve 3 are 0.80, 1.00 and 1.20 respectively, and other parameters are consistent, the early warning data accuracy assessment index of curve 1, curve 2 and curve 3 are 0.89, 0.94 and 0.98 respectively. It can be seen that when the eye state assessment data accuracy index is consistent with other parameters, when the physiological state assessment data accuracy index is larger, the early warning data accuracy assessment index is larger.

[0074] Among them, the specific process of analyzing the accuracy of the warning data and issuing a warning based on the warning data accuracy evaluation index is as follows: when the warning data accuracy evaluation index is within the normal range, it means that the obtained warning data is accurate and the obtained warning data can be used for warning; when the warning data accuracy evaluation index is within the abnormal range, it means that the obtained warning data is inaccurate, and the warning data will not be used to warn the fatigue status of the operator.

[0075] In this embodiment, when the warning data accuracy index is within the normal range, it indicates that the warning data of the operator acquired at the same time is relatively accurate. In this case, the average value of the corresponding data of the two sets of eye state assessment data and the two sets of physiological state assessment data in the warning data can be taken as the warning data used in the warning process to assess the fatigue state of the operator. When the warning data accuracy index is within the abnormal range, it indicates that the warning data of the operator acquired at the same time is inaccurate, and there are certain problems with the obtained warning data. In this case, the relatively better set of eye state assessment data of the two sets can be selected as the eye state assessment data used in the warning of the operator's fatigue state based on the image quality analysis index. Similarly, the relatively better set of physiological state assessment data of the two sets can be selected as the physiological state assessment data used in the warning of the operator's fatigue state based on the physiological state assessment data selection index. At the same time, the next set of warning data is continuously acquired to assess the accuracy of the warning data. If the warning data accuracy index is within the abnormal range for five consecutive times, it indicates that the equipment acquiring the warning data has an abnormality and the equipment should be promptly inspected and replaced.

[0076] In this embodiment, the main process of issuing an early warning for the fatigue state of the operator is: collecting the initial state data of the operator; obtaining the eye state evaluation data of the operator, obtaining the eye fatigue state evaluation index based on the eye state evaluation data and the initial state data, analyzing the eye fatigue state of the operator according to the eye fatigue state evaluation index and issuing a first early warning; obtaining the physiological state evaluation data of the operator, obtaining the physiological fatigue state evaluation index of the operator according to the physiological state evaluation data and the initial state data, analyzing the physiological fatigue state of the operator according to the physiological fatigue state evaluation index and issuing a second early warning; obtaining the comprehensive evaluation index of the operator's fatigue state by combining the eye fatigue state evaluation index and the physiological fatigue state evaluation index; analyzing the overall fatigue state of the operator according to the comprehensive evaluation index of the operator's fatigue state and issuing three early warnings.

[0077] In this embodiment, the specific method for obtaining the eye fatigue state evaluation index is: obtaining the weight ratios of the preset blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure duration in the eye fatigue state evaluation index from the database;

[0078]

[0079] In the formula, β is the eye fatigue status evaluation index, f is the blink frequency of the operator, t is the blink duration of the operator, s is the eye movement speed of the operator, d is the pupil diameter of the operator, T is the eyelid closure duration of the operator, f1 is the initial blink frequency of the operator, t1 is the initial blink duration of the operator, s1 is the initial eye movement speed of the operator, d1 is the initial pupil diameter of the operator, T1 is the initial eyelid closure duration of the operator, and b1 is b2 represents the weight ratio of blink frequency in the eye fatigue status evaluation index, b3 represents the weight ratio of eye movement speed in the eye fatigue status evaluation index, b4 represents the weight ratio of pupil diameter in the eye fatigue status evaluation index, b5 represents the weight ratio of eyelid closure time in the eye fatigue status evaluation index, e is a natural constant. It should be noted that the data used in the above formula is accurate data obtained through analysis of eye status evaluation data accuracy indicators.

[0080] In this embodiment, the initial blinking frequency, the initial blinking duration, the initial eye movement speed, the initial pupil diameter, the initial eyelid closure duration, the initial heart rate, the initial skin conductivity, the initial number of EEG alpha waves and the initial blood oxygen saturation are the initial state data of the operator. Obtaining the data of the operator in the initial awake state when performing the operation is equivalent to obtaining the initial state data of the operator. The initial heart rate can be obtained by placing an optical sensor on the steering wheel or seat, the initial skin conductivity can be obtained by installing a non-contact sensor, such as a capacitive sensor, on the steering wheel or seat, the initial number of EEG alpha waves can be obtained by wireless dry electrode technology, and the initial blood oxygen saturation can be obtained by a pulse oximeter clipped to the earlobe; the initial blinking frequency, the initial blinking duration, the initial eye movement speed, the initial pupil diameter and the initial eyelid closure duration are obtained by an eye tracker.

[0081] In this embodiment, f represents the number of blinks per unit time, t represents the time length from the start to the end of a blink, s represents the speed of the eyeball moving in the field of view, d represents the size of the pupil diameter, T represents the length of time the eyelid is completely closed, f1 represents the number of blinks per unit time in the initial state of the operator, t1 represents the time length from the start to the end of a blink in the initial state of the operator, s1 represents the speed of the eyeball moving in the field of view in the initial state, d1 represents the size of the pupil diameter in the initial state of the operator, and T1 represents the length of time the eyelid is completely closed in the initial state of the operator.

[0082] In this embodiment, b1, b2, b3, b4 and b5 are respectively the influence weight ratios of the preset blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure time obtained from the preset database, which are used to represent the numerical values ​​of the influence degree of blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure time on the eye state assessment data. When used, the influence weight ratios of the blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure time can be directly obtained from the preset database, and the corresponding relationship can be A mapping relationship is pre-set. For example, blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure time and their corresponding weights form a mapping set. After the real-time blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure time are input into the mapping set, the corresponding weight ratios of blink frequency, blink duration, eye movement speed, pupil diameter and eyelid closure time will be obtained. In this example, the value ranges of b1, b2, b3, b4 and b5 are all [0, 1], and the sum of b1, b2, b3, b4 and b5 is 1.

[0083] In this embodiment, the blinking frequency, blinking duration, eye movement speed, pupil diameter and eyelid closure time are comprehensively used to evaluate the eye fatigue state of the operator, and the eye fatigue state of the operator is obtained so as to issue an early warning. The closer the blinking frequency, blinking duration, eye movement speed, pupil diameter and eyelid closure time are to the initial blinking frequency, initial blinking duration, initial eye movement speed, initial pupil diameter and initial eyelid closure time, the more awake the operator is. The slower the blinking frequency, the longer the blinking duration and eyelid closure time, and the slower the eye movement speed. This embodiment uses logarithmic operation for nonlinear transformation to smooth the influence of outliers, which is conducive to improving the accuracy and effectiveness of eye fatigue state evaluation indicators.

[0084] In this embodiment, the preset eye fatigue state evaluation index threshold value obtained is β1, β1=1.07. When the eye fatigue state evaluation index is less than or equal to the eye fatigue state evaluation index threshold value, that is, β≤β1, it indicates that the eye fatigue state evaluation index is in the normal range, that is, it indicates that the eye fatigue state of the operator is relatively mild or the operator's eyes are not in a fatigue state; when it is in other situations, that is, β>β1, it indicates that the eye fatigue state evaluation index is in an abnormal range, that is, it indicates that the operator's eyes are in a fatigue state.

[0085] In this embodiment, when the eye fatigue status evaluation index is within the normal range, it indicates that the operator is relatively awake and no warning is issued; when the eye fatigue status evaluation index is within the abnormal range, it indicates that the operator's eyes are in a relatively fatigued state. At this time, the operator is given a first warning, and an obvious indicator light is used to remind the operator that the eyes are in a fatigued state. The operator is reminded to pay attention to eye rest through voice, and the operator is advised to relax his eyes appropriately in a place where he can park.

[0086] In this embodiment, the specific method for obtaining the physiological fatigue state evaluation index is as follows: obtaining from a database the preset eye fatigue state evaluation index, heart rate, skin conductivity, number of alpha waves in the brain wave, and weight ratio of blood oxygen saturation in the physiological fatigue state evaluation index;

[0087]

[0088] In the formula, δ is the physiological fatigue state evaluation index, β is the eye fatigue state evaluation index, a is the heart rate of the operator, c is the skin conductivity of the operator, y is the number of alpha waves in the operator's brain wave, u is the blood oxygen saturation of the operator, a1 is the initial heart rate of the operator, c1 is the initial skin conductivity of the operator, y1 is the initial number of alpha waves in the operator's brain wave, u1 is the initial blood oxygen saturation of the operator, d1 is the weight ratio of heart rate in the physiological fatigue state evaluation index, d2 is the weight ratio of heart rate in the physiological fatigue state evaluation index, d3 is the weight ratio of heart rate in the physiological fatigue state evaluation index, d4 is the weight ratio of heart rate in the physiological fatigue state evaluation index, d5 is the weight ratio of heart rate in the physiological fatigue state evaluation index, and e is a natural constant. It should be noted that the data used in the above formula are accurate data obtained through analysis of the accuracy index of physiological state evaluation data.

[0089] In this embodiment, a represents the number of heart beats per unit time, c represents the electrical conductivity of the skin surface, y represents the number of alpha waves in the electrical activity of cerebral cortical nerve cells, u represents the percentage of oxygen content in hemoglobin in the blood, a1 represents the number of heart beats per unit time in the initial state of the operator, c1 represents the electrical conductivity of the skin surface in the initial state of the operator, y1 represents the number of alpha waves in the electrical activity of cerebral cortical nerve cells in the initial state of the operator, and u1 represents the percentage of oxygen content in the hemoglobin in the blood in the initial state of the operator.

[0090] In this embodiment, d1, d2, d3, d4 and d5 are respectively the influence weight ratios of the preset eye fatigue status evaluation index, heart rate, skin conductivity, number of brain wave alpha waves and blood oxygen saturation obtained from the preset database, which are used to represent the numerical values ​​of the degree of influence of the eye fatigue status evaluation index, heart rate, skin conductivity, number of brain wave alpha waves and blood oxygen saturation on the eye status evaluation data. When used, the influence weight ratios of the eye fatigue status evaluation index, heart rate, skin conductivity, number of brain wave alpha waves and blood oxygen saturation can be directly obtained from the preset database, and the corresponding relationship can be This is a pre-set mapping relationship. For example, the eye fatigue status assessment index, heart rate, skin conductivity, the number of brain wave alpha waves and blood oxygen saturation and their corresponding weights form a mapping set. After the real-time eye fatigue status assessment index, heart rate, skin conductivity, the number of brain wave alpha waves and blood oxygen saturation are input into the mapping set, the corresponding eye fatigue status assessment index, heart rate, skin conductivity, the number of brain wave alpha waves and blood oxygen saturation will be obtained. The value range of d1, d2, d3, d4 and d5 in this example is [0, 1], and the sum of d1, d2, d3, d4 and d5 is 1.

[0091] In this embodiment, the fatigue degree of the operator's physiological state is evaluated by comprehensively considering the heart rate, skin conductivity, number of alpha waves in the EEG, and blood oxygen saturation, and the physiological state of the operator is accurately evaluated. The closer the heart rate, skin conductivity, number of alpha waves in the EEG, and blood oxygen saturation are to the initial heart rate, initial skin conductivity, initial number of alpha waves in the EEG, and initial blood oxygen saturation, the closer the operator's physiological state is to the physiological state when awake. The smaller the eye fatigue state evaluation index is, the closer the heart rate, skin conductivity, number of alpha waves in the EEG, and blood oxygen saturation are to the corresponding data when the operator is in the initial state. This embodiment uses exponential operation for nonlinear transformation, so that the obtained physiological fatigue state evaluation index is more explanatory.

[0092] In this embodiment, the preset physiological fatigue state evaluation index threshold value obtained is δ1, δ1=1.11. When the physiological fatigue state evaluation index is less than or equal to the physiological fatigue state evaluation index threshold value, that is, δ≤δ1, it indicates that the physiological fatigue state evaluation index is in the normal range, that is, the physiological state of the operator is relatively awake; when in other situations, that is, δ>δ1, it indicates that the physiological fatigue state evaluation index is in the abnormal range, that is, the physiological state of the operator is relatively tired.

[0093] In this embodiment, when the physiological fatigue state evaluation index is within the normal range, it means that the physiological state of the operator is closer to the physiological state when awake, and no secondary warning is issued. When the physiological fatigue state evaluation index is within the abnormal range, it means that the physiological state of the operator is in a fatigue state. At this time, a secondary warning is issued to the operator, and a voice warning is given to the operator that the physiological state is in a fatigue state, warning the operator that the physical condition is in a relatively fatigued state. At this time, the fatigue state of the operator is already greater than the fatigue state during the first warning, and it is recommended to replace the operator or take a rest in time.

[0094] In this embodiment, the specific method for obtaining the comprehensive evaluation index of the fatigue state of the operator is: obtaining the weight ratio of the preset eye fatigue state evaluation index and the physiological fatigue state evaluation index in the comprehensive evaluation index of the fatigue state of the operator from the database;

[0095]

[0096] Wherein, θ is the comprehensive evaluation index of the fatigue state of the operator, β is the evaluation index of the eye fatigue state, δ is the evaluation index of the physiological fatigue state, r1 is the weight ratio of the eye fatigue state evaluation index in the comprehensive evaluation index of the fatigue state of the operator, r2 is the weight ratio of the physiological fatigue state evaluation index in the comprehensive evaluation index of the fatigue state of the operator, and e is a natural constant.

[0097] In this embodiment, r1 and r2 are respectively the influence weight ratios corresponding to the preset eye fatigue state assessment index and the physiological fatigue state assessment index obtained from the preset database, which are used to represent the numerical values ​​of the degree of influence of the eye fatigue state assessment index and the physiological fatigue state assessment index on the eye state assessment data. When used, the influence weight ratios corresponding to the eye fatigue state assessment index and the physiological fatigue state assessment index can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the eye fatigue state assessment index and the physiological fatigue state assessment index and their corresponding weights form a mapping set. After the real-time eye fatigue state assessment index and the physiological fatigue state assessment index are input into the mapping set, the corresponding weight ratios corresponding to the corresponding eye fatigue state assessment index and the physiological fatigue state assessment index will be obtained. In this example, the value ranges of r1 and r2 are both [0, 1], and the sum of r1 and r2 is 1.

[0098] In this embodiment, the eye fatigue state evaluation index and the physiological fatigue state evaluation index are combined to perform a comprehensive evaluation of the fatigue state of the operator. When the comprehensive evaluation index of the operator's fatigue state is in an abnormal range, it means that the operator is in a fatigue state, which will have a certain impact on the operator's judgment and operation. The larger the eye fatigue state evaluation index and the physiological fatigue state evaluation index, the larger the comprehensive evaluation index of the operator's fatigue state, that is, the greater the degree of fatigue of the operator. This embodiment uses the inverse sine function for nonlinear transformation, so that the obtained comprehensive evaluation index of the operator's fatigue state is more reliable and effective.

[0099] In this embodiment, the preset threshold value of the comprehensive evaluation index of the fatigue state of the operator is θ1, θ1=1.77. When the comprehensive evaluation index of the fatigue state of the operator is less than or equal to the threshold value of the comprehensive evaluation index of the fatigue state of the operator, that is, θ≤θ1, it means that the comprehensive evaluation index of the fatigue state of the operator is in the normal range, that is, the operator's fatigue state is relatively mild; when it is in other situations, that is, θ>θ1, it means that the comprehensive evaluation index of the fatigue state of the operator is in the abnormal range, that is, the operator's fatigue state is relatively serious.

[0100] like Figure 3 As shown, it is a graph of comprehensive evaluation indexes of fatigue status of operators provided by an embodiment of the present invention, with eye fatigue status evaluation index as the independent variable and comprehensive evaluation index of fatigue status of operators as the dependent variable. Through the increase and decrease analysis of curve 1, curve 2 and curve 3, it can be seen that when the eye fatigue status evaluation index is larger, the comprehensive evaluation index of fatigue status of operators is larger; when the eye fatigue status evaluation index is smaller, the comprehensive evaluation index of fatigue status of operators is smaller. The statistical table of relevant data of comprehensive evaluation index of fatigue status of operators is shown in Table 2:

[0101] Table 2 Statistics of comprehensive evaluation indicators of fatigue status of operators

[0102]

[0103] It can be seen from Table 1 that for curve 1, curve 2 and curve 3, when the eye fatigue state evaluation index of curve 1, curve 2 and curve 3 is 1.00, when the physiological fatigue state evaluation index of curve 1, curve 2 and curve 3 is 0.90, 1.00 and 1.10 respectively, and other parameters are consistent, the comprehensive evaluation index of the operator fatigue state of curve 1, curve 2 and curve 3 is 1.40, 1.70 and 2.00 respectively. It can be seen that when the eye fatigue state evaluation index is consistent with other parameters, when the physiological fatigue state evaluation index is larger, the comprehensive evaluation index of the operator fatigue state is larger.

[0104] In this embodiment, when the comprehensive evaluation index of the operator's fatigue state is within the normal range, it indicates that the operator's fatigue state is relatively light, and three warnings are not required at this time. When the comprehensive evaluation index of the operator's fatigue state is within the abnormal range, it indicates that the operator's fatigue level is relatively heavy. At this time, the operator's judgment and operation may have problems. At this time, three warnings are required for the operator. At this time, the operator is already in an extremely fatigued state. The operator's continued operation may affect the operator's life and property safety. The operator needs to be seriously warned, and the operator needs to be given continuous voice warnings. The operator should be kept awake by lowering the temperature in the car or opening the window, and compulsory measures should be taken to limit the operator's driving speed. The operator is warned by voice to end the trip quickly and rest in time.

[0105] like Figure 4 As shown, a flow chart of an early warning method for avoiding fatigue work provided by an embodiment of the present invention includes the following steps: collecting early warning data obtained by two groups of identical equipment at two preset positions at the same time by the operator; obtaining two groups of eye state assessment data and image data in the early warning data and analyzing them to obtain an eye state assessment data accuracy index and an image quality analysis index, analyzing the accuracy of the two groups of eye state assessment data according to the eye state assessment data accuracy index, and selecting accurate data for eye state assessment according to the image quality analysis index; obtaining two groups of physiological state assessment data in the early warning data and analyzing them to obtain a physiological state assessment data accuracy index and a physiological state assessment data selection index, analyzing the accuracy of the two groups of physiological state assessment data according to the physiological state assessment data accuracy index, and determining valid data for physiological state assessment according to the physiological state assessment data selection index; comprehensively obtaining the eye state assessment data accuracy index and the physiological state assessment data accuracy index to obtain the operator's early warning data accuracy assessment index; analyzing the accuracy of the warning data according to the early warning data accuracy assessment index and issuing an early warning.

[0106] In summary, the embodiment of the present invention obtains two sets of eye state assessment data of the operator, thereby obtaining the accuracy index of the eye state assessment data of the operator and evaluating the two sets of eye state assessment data of the operator, and obtains two sets of eye state image data for analysis to obtain image quality analysis index, and selects valid data according to the image quality analysis index, so as to obtain accurate data in a timely and effective manner, and then obtains two sets of physiological state assessment data of the operator, obtains and determines the accurate data for eye state assessment according to the physiological state assessment data selection index, thereby obtaining the accuracy index of the physiological state assessment data of the operator and analyzing the two sets of physiological state assessment data of the operator, and finally comprehensively obtains the eye state assessment data accuracy index and the physiological state assessment data accuracy index to obtain the warning data accuracy assessment index, and accurately analyzes the accuracy of the warning data according to the warning data accuracy assessment index, and then issues a warning based on the accurate warning data, which effectively solves the problem in the existing technology that it is difficult to accurately and timely obtain effective warning data for the operator.

[0107] There are a few points to note:

[0108] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0109] (2) For the sake of clarity, the thickness of layers or regions in the drawings used to describe the embodiments of the present invention are exaggerated or reduced, that is, these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or intervening elements may be present.

[0110] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to form new embodiments.

[0111] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A warning system for avoiding fatigue work, characterized in that: include: Data collection module, eye status data accuracy assessment module, physiological status data accuracy assessment module, early warning data accuracy assessment module and analysis module; The data collection module collects warning data obtained by operators at two preset locations at the same time using two sets of identical equipment; The early warning data includes two sets of eye status assessment data; The two sets of eye state assessment data include: a first set of eye state assessment data and a second set of eye state assessment data, the first set of eye state assessment data including a first set of blink frequency, a first set of blink duration, a first set of eye movement velocity, a first set of pupil diameter, and a first set of eyelid closure duration data, and the second set of eye state assessment data including a second set of blink frequency, a second set of blink duration, a second set of eye movement velocity, a second set of pupil diameter, and a second set of eyelid closure duration data; The eye state data accuracy assessment module obtains two sets of eye state assessment data and eye state image data in the early warning data and analyzes them to obtain an eye state assessment data accuracy index and an image quality analysis index, analyzes the accuracy of the two sets of eye state assessment data according to the eye state assessment data accuracy index, and selects valid data for eye state assessment according to the image quality analysis index; The specific method for obtaining the accuracy index of the eye status assessment data is as follows: Where α is the accuracy index of eye status assessment data, f ” Expressed as the second group of blink frequency, t ” Expressed as the duration of the second group of blinks, s ” Expressed as the second set of eye movement velocity, d ” Expressed as the second set of pupil diameters, T ” Expressed as the duration of eyelid closure for the second group, f ' Expressed as the first group of blink frequency, t ' Expressed as the duration of the first group of blinks, s ' Expressed as the first set of eye movement velocity, d ' Expressed as the first group of pupil diameters, T ' It is expressed as the duration of eyelid closure of the first group, and e is a natural constant; The physiological state data accuracy assessment module obtains two sets of physiological state assessment data from the early warning data and analyzes them to obtain a physiological state assessment data accuracy index and a physiological state assessment data selection index, analyzes the accuracy of the two sets of physiological state assessment data according to the physiological state assessment data accuracy index, and determines valid data for physiological state assessment according to the physiological state assessment data selection index; The early warning data accuracy assessment module: comprehensively calculates the eye state assessment data accuracy index and the physiological state assessment data accuracy index to obtain the operator's early warning data accuracy assessment index; The analysis module analyzes the accuracy of the warning data and issues a warning based on the warning data accuracy evaluation index.

2. The early warning system for avoiding fatigue work according to claim 1, characterized in that: The early warning data includes two sets of physiological status assessment data; The two groups of physiological state assessment data include a first group of physiological state assessment data and a second group of physiological state assessment data, the first group of physiological state assessment data includes a first group of heart rate, a first group of skin conductivity, a first group of brain wave alpha wave number and a first group of blood oxygen saturation, and the second group of physiological state assessment data includes a second group of heart rate, a second group of skin conductivity, a second group of brain wave alpha wave number and a second group of blood oxygen saturation.

3. The early warning system for avoiding fatigue work according to claim 1, characterized in that: The specific analysis process of the eye state assessment data accuracy index is as follows: obtaining a preset eye state assessment data accuracy index threshold, obtaining the eye state assessment data accuracy index, and comparing the obtained eye state assessment data accuracy index with the eye state assessment data accuracy index threshold; when the eye state assessment data accuracy index is greater than the eye state assessment data accuracy index threshold, the eye state assessment data accuracy index is in an abnormal range; When the eye condition assessment data accuracy index is less than or equal to the eye condition assessment data accuracy index threshold, the eye condition assessment data accuracy index is within a normal range.

4. The early warning system for avoiding fatigue work according to claim 3, characterized in that: The specific process of analyzing the accuracy of the two sets of eye state assessment data according to the eye state assessment data accuracy index is as follows: when the eye state assessment data accuracy index is within the normal range, it means that the two sets of eye state assessment data obtained are accurate, and the two sets of eye state assessment data at this time can be used to assess the eye state and issue an early warning; when the eye state assessment data accuracy index is within the abnormal range, it means that the two sets of eye state assessment data obtained at this time are inaccurate; when the eye state assessment data accuracy index is within the abnormal range, the eye state image data is obtained for analysis to obtain the image quality analysis index, and the valid data for eye state assessment is determined according to the image quality analysis index.

5. The early warning system for avoiding fatigue work according to claim 1, characterized in that: The specific analysis process of the physiological state assessment data accuracy index is as follows: obtaining a preset physiological state assessment data accuracy index threshold, obtaining the physiological state assessment data accuracy index, and comparing the obtained physiological state assessment data accuracy index with the physiological state assessment data accuracy index threshold; when the physiological state assessment data accuracy index is greater than the physiological state assessment data accuracy index threshold, the physiological state assessment data accuracy index is in an abnormal range; When the physiological state assessment data accuracy index is less than or equal to the physiological state assessment data accuracy index threshold, the physiological state assessment data accuracy index is within a normal range.

6. The early warning system for avoiding fatigue work according to claim 5, characterized in that: The specific process of analyzing the accuracy of the two sets of physiological state assessment data according to the physiological state assessment data accuracy index is as follows: when the physiological state assessment data accuracy index is within the normal range, it means that the two sets of physiological state assessment data obtained are accurate; when the physiological state assessment data accuracy index is within the abnormal range, it means that the two sets of physiological state assessment data obtained are inaccurate; when the physiological state assessment data accuracy index is within the abnormal range, the physiological state assessment data selection index is obtained, and the valid data for physiological state assessment is determined according to the physiological state assessment data selection index.

7. The early warning system for avoiding fatigue work according to claim 1, characterized in that: The specific analysis process of the early warning data accuracy evaluation indicator is as follows: obtaining a preset early warning data accuracy evaluation indicator threshold, obtaining the early warning data accuracy evaluation indicator, comparing the obtained early warning data accuracy evaluation indicator with the early warning data accuracy evaluation indicator threshold, and when the early warning data accuracy evaluation indicator is greater than the early warning data accuracy evaluation indicator threshold, the early warning data accuracy evaluation indicator is in an abnormal range; When the early warning data accuracy evaluation index is less than or equal to the early warning data accuracy evaluation index threshold, the early warning data accuracy evaluation index is in a normal range.

8. The early warning system for avoiding fatigue work according to claim 7, characterized in that: The specific process of analyzing the accuracy of the warning data and issuing a warning based on the warning data accuracy evaluation index is as follows: when the warning data accuracy evaluation index is within the normal range, it means that the obtained warning data is accurate and the obtained warning data can be used for warning; when the warning data accuracy evaluation index is within the abnormal range, it means that the obtained warning data is inaccurate and the warning data will not be used to warn the fatigue status of the operator.

9. An early warning method for avoiding fatigue work, the early warning method is based on the early warning system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Collect early warning data obtained by operators at two preset locations at the same time using two sets of identical equipment; Obtaining two sets of eye state assessment data and image data from the early warning data and analyzing them to obtain an eye state assessment data accuracy index and an image quality analysis index, analyzing the accuracy of the two sets of eye state assessment data according to the eye state assessment data accuracy index, and selecting accurate data for eye state assessment according to the image quality analysis index; Obtaining two sets of physiological state assessment data from the early warning data and analyzing them to obtain a physiological state assessment data accuracy index and a physiological state assessment data selection index, analyzing the accuracy of the two sets of physiological state assessment data based on the physiological state assessment data accuracy index, and determining valid data for physiological state assessment based on the physiological state assessment data selection index; The accuracy index of the eye status assessment data and the physiological status assessment data are combined to obtain the accuracy index of the operator's early warning data; Analyze the accuracy of warning data and issue warnings based on the warning data accuracy evaluation indicators.

Images