Fatigue driving adaptive detection method

Abstract

The invention provides a fatigue driving adaptive detection method, which comprises the following steps: step S101, acquiring facial features of a driver during driving, and calculating a fatigue index of the driver according to the facial features of the driver and a fatigue monitoring calculation method; s102, when the calculated fatigue index exceeds a set fatigue index normal threshold value, a fatigue driving signal is sent to the vehicle; s103, if the driver does not have the fatigue driving behavior, re-judging the type to which the eyes of the driver belong; step S104, modifying parameters in a fatigue monitoring calculation method according to the type of the eyes of the driver, and performing adaptive adjustment on the fatigue monitoring calculation method; and S105, circulating the above steps by using the adjusted fatigue monitoring calculation method until the monitoring is normal. According to the invention, the problem of false alarm of the fatigue monitoring system caused by small eyes of the driver is solved, and the accuracy and adaptability of fatigue monitoring are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle fatigue monitoring, and more specifically to a fatigue driving adaptive detection method. Background Art

[0002] In modern society, with the rapid development of the automobile industry, intelligent driving technology has become a key way to improve road safety and optimize driving experience. Among them, the driver monitoring system, as a core component of the intelligent driving assistance system, is of immeasurable importance in preventing traffic accidents and ensuring the safety of drivers and passengers. Especially in long-distance or monotonous driving environments, drivers are very likely to feel fatigued, and fatigue driving is one of the main causes of frequent traffic accidents. Therefore, it is particularly urgent and important to develop an efficient and accurate driver fatigue monitoring system.

[0003] Among the many fatigue monitoring technologies, the vision-based driver fatigue monitoring system has attracted much attention due to its non-contact, real-time feedback and relatively low cost. The system uses a camera to capture the driver's facial image, and then uses advanced computer vision and image processing technology to accurately extract facial features and conduct in-depth analysis of the driver's fatigue status. Among them, eye and lip features have become the main focus of monitoring due to their high correlation with fatigue status.

[0004] However, despite significant technological progress, vision-based driver fatigue monitoring systems still face some challenges that need to be addressed. In particular, for people with small eyes, traditional eye feature extraction and fatigue determination algorithms often have a high risk of false alarms. The reason is that when small eyes are closed or in a fatigued state, the changes in their visual features are not as significant as those of large eyes, which makes it difficult for the algorithm to effectively distinguish between normal blinking and closing of eyes due to fatigue, which may falsely trigger alarms and cause unnecessary trouble to some drivers with small eyes.

[0005] Currently, the existing technologies on the market have not fully taken into account the adaptive adjustments for people with small eyes, which leads to some drivers frequently encountering false alarms from fatigue monitoring systems due to the size of their eyes. Summary of the invention

[0006] The present invention provides a fatigue driving adaptive detection method, which aims to solve the problem of false alarm of a fatigue monitoring system caused by the small eyes of the driver himself, and improve the accuracy and adaptability of fatigue monitoring.

[0007] To achieve the above object, the present invention provides a method for adaptively detecting fatigue driving, comprising:

[0008] Step S101, acquiring facial features of the driver while driving, and calculating the driver's fatigue index according to the driver's facial features and a fatigue monitoring calculation method;

[0009] Step S102, when the calculated fatigue index exceeds the set fatigue index normal threshold, sending a fatigue driving signal to the vehicle;

[0010] Step S103, if the driver does not have fatigue driving behavior, re-determine the driver's eye type;

[0011] Step S104, modifying the parameters in the fatigue monitoring calculation method according to the type of the driver's eyes, and adaptively adjusting the fatigue monitoring calculation method;

[0012] Step S105, using the adjusted fatigue monitoring calculation method to cycle through the above operations until the monitoring is normal.

[0013] In one embodiment, obtaining the facial features of the driver while driving specifically includes:

[0014] A video captured in real time while the driver is driving is obtained, and facial feature points of the driver are extracted from continuous video frames, wherein the facial feature points of the driver include eye feature points and lip feature points, and the eye aspect ratio and the lip aspect ratio are calculated based on the eye feature points and the lip feature points.

[0015] In one embodiment, the eye feature points of the driver include: the leftmost end of the left eye, the rightmost end of the left eye, the upper end of the left eye, the lower end of the left eye, the leftmost end of the right eye, the rightmost end of the right eye, the upper end of the right eye, and the lower end of the right eye;

[0016] The driver's lip feature points include: the leftmost end of the lip, the rightmost end of the lip, the upper end of the lip, and the lower end of the lip.

[0017] In one embodiment, the driver's eye aspect ratio calculation formula is specifically:

[0018]

[0019]

[0020]

[0021] Among them, EAR L EAR R are the aspect ratios of the driver's left eye and right eye respectively, EAR is the driver's average eye aspect ratio, and p1, p2, p3, p4, p5, p6, p7, p8, p9, p10, p11, and p12 are the coordinate information of the eye feature points.

[0022] In one embodiment, the driver's lip aspect ratio calculation formula is specifically:

[0023]

[0024] Among them, MAR is the aspect ratio of the driver's lips, and p13, p14, p15, p16, p17, and p18 are the coordinate information of the lip feature points.

[0025] In one embodiment, the eye aspect ratio is set with an eye aspect ratio threshold EAR threshold , based on the comparison between the driver's eye aspect ratio and the eye aspect ratio threshold, the number of eye closures and eye closure time are calculated, and then the eye confidence is calculated. eye ;

[0026] The lip aspect ratio is set with a lip aspect ratio threshold MAR threshold According to the driver's lip aspect ratio and lip aspect ratio threshold, the number of yawns and yawning time are calculated, and then the lip confidence is calculated. mouth .

[0027] In one embodiment, the eye confidence calculation specifically includes: counting each video frame of the driver while driving, and when EAR < EAR threshold When the eye is closed, it is judged as open. The closed eye time is calculated by the number of closed eye times in consecutive video frames, and then the eye confidence in the fatigue monitoring algorithm is calculated by the percentage of closed eye time in the total time. eye ;

[0028] The lip confidence calculation specifically includes: counting each video frame when the driver is driving, and when MAR<MAR tnreshold When the mouth is closed, it is judged as open. The mouth opening time is calculated by the number of mouth opening times in consecutive video frames, and then the lip confidence in the fatigue monitoring algorithm is calculated by the percentage of mouth opening time in the total time. mouth .

[0029] In one embodiment, the specific algorithm of the fatigue indicator in the fatigue monitoring algorithm is:

[0030] Confidence=a*Confidence eye +b*Confidence mout

[0031] Wherein, a and b are the proportion parameters of the set eye confidence and lip confidence in the total confidence, respectively, and a+b=1.

[0032] In one embodiment, the driver's eye type includes a first eye type and a second eye type, and the driver's eye type is provided with an eye morphology recognition threshold;

[0033] When the eye aspect ratio of the left eye and the right eye of the driver is greater than the set eye morphology recognition threshold when the driver is not fatigued, the driver's eye type belongs to the second eye type; otherwise, it belongs to the first eye type.

[0034] In one embodiment, when the driver's eye type belongs to the first eye type, the eye confidence ratio parameter a in the fatigue monitoring algorithm is reduced;

[0035] When the driver belongs to the second eye type, the parameter a of the eye confidence ratio in the fatigue monitoring algorithm is increased.

[0036] The present invention has the following beneficial effects:

[0037] Adaptability: This method automatically adjusts the eye confidence ratio parameter in the fatigue detection algorithm according to the driver's eye type, thereby increasing the flexibility and personalization of fatigue monitoring, making the calculation results more in line with the actual situation of each driver and reducing the false alarm rate caused by individual differences.

[0038] Accuracy: This method not only analyzes the driver's eye features, but also incorporates multiple fatigue indicators such as lip features. By calculating various fatigue indicators in real time, it can more comprehensively evaluate the driver's fatigue state, greatly improving the accuracy of detection.

[0039] Safety: This method does not require additional collection of driver facial feature training models, so it does not require high computing resources and algorithm support, which ensures the driver's privacy and data security and achieves more accurate and reliable driver fatigue monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of a process flow of a method for adaptively detecting fatigue driving according to an embodiment of the present invention;

[0041] Figure 2 A schematic diagram of selecting facial feature points in a method for adaptively detecting fatigue driving according to an embodiment of the present invention;

[0042] Figure 3 A schematic diagram of signal transmission for a driver to judge the eye size by himself in a method for adaptively detecting fatigue driving according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the embodiment of this application will be described in more detail below in conjunction with the drawings in the embodiment of this application. In the drawings, the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of this application, not all of them. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain this application, and should not be construed as limitations on this application. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0044] Figure 1 The following is a flow chart of a method for adaptively detecting fatigue driving according to an embodiment of the present invention. The method for adaptively detecting fatigue driving comprises:

[0045] Step S101, acquiring facial features of the driver while driving, and calculating the driver's fatigue index according to the driver's facial features and a fatigue monitoring calculation method;

[0046] Step S102, when the calculated fatigue index exceeds the set fatigue index normal threshold, sending a fatigue driving signal to the vehicle;

[0047] Step S103, if the driver does not have fatigue driving behavior, re-determine the driver's eye type;

[0048] Step S104, modifying the parameters in the fatigue monitoring calculation method according to the type of the driver's eyes, and adaptively adjusting the fatigue monitoring calculation method;

[0049] Step S105, using the adjusted fatigue monitoring calculation method to cycle through the above operations until the monitoring is normal.

[0050] Exemplarily, when the fatigue index calculated by the fatigue monitoring calculation method exceeds the normal threshold of the fatigue index, the fatigue monitoring system will send a signal value of driver_is_fatigue_status=1 to the vehicle computer. When the vehicle computer receives this signal, it will remind the driver not to drive while fatigued through images, text and sound.

[0051] Specifically, when the calculated fatigue index does not exceed the set normal threshold of fatigue index, the facial features of the driver while driving continue to be counted.

[0052] When the driver is driving fatigued, he can choose to rest at a nearby place.

[0053] When the driver is driving fatigued and does not stop, the system will continue to send out fatigue driving signals. The driver can choose to turn off the fatigue driving reminder function or not.

[0054] When the driver is not driving fatigued, the driver's eye type can be reselected.

[0055] Furthermore, the re-determination of the driver's eye type may be automatic identification by the fatigue monitoring system or active selection of the driver's eye type.

[0056] For example, when the vehicle stops and safety is guaranteed, the vehicle terminal will pop up a pop-up window, such as Figure 3 As shown, the driver is allowed to set or confirm whether he belongs to the small-eye group according to his own situation. After the user confirms, the vehicle computer will output the signal driver_small_eye_status=1 to the fatigue monitoring system, and then automatically adjust the confidence threshold of the number of closed eyes and the proportion of closed eye time in the fatigue detection algorithm. eye If this confidence level meets the driver's needs or the driver determines that he or she does not belong to the group of people with small eyes, the driver_small_eye_status signal value will be changed to 0. You can choose not to modify the parameter.

[0057] If this confidence does not meet the driver's needs and it is determined that the driver belongs to the group of people with small eyes, this step will be looped until the eye confidence parameter reaches the minimum threshold.

[0058] When the driver is prompted that the eye confidence has reached the algorithm's minimum threshold, it is recommended that the driver turn off the fatigue driving detection function or simply cancel eye monitoring.

[0059] In one embodiment, obtaining the facial features of the driver while driving specifically includes:

[0060] A video captured in real time while the driver is driving is obtained, and facial feature points of the driver are extracted from continuous video frames, wherein the facial feature points of the driver include eye feature points and lip feature points, and the eye aspect ratio and the lip aspect ratio are calculated based on the eye feature points and the lip feature points.

[0061] Specifically, the facial features may further include head facial features, and the head facial features may be used to calculate the degree of head bowing and the degree of inclination of the driver's head posture.

[0062] In one embodiment, the eye feature points of the driver include: the leftmost end of the left eye, the rightmost end of the left eye, the upper end of the left eye, the lower end of the left eye, the leftmost end of the right eye, the rightmost end of the right eye, the upper end of the right eye, and the lower end of the right eye;

[0063] The driver's lip feature points include: the leftmost end of the lip, the rightmost end of the lip, the upper end of the lip, and the lower end of the lip.

[0064] Specifically, two or more points may be selected at the upper end of the left eye to better analyze the eye features of the driver.

[0065] Furthermore, the selection method of the lower end of the eye, the upper end of the lip, and the lower end of the lip is similar to the selection method of the upper end of the left eye.

[0066] Exemplarily, when the driver's head feature is set with head feature points, the head feature points can be set at the center of the eyebrows and the chin, and the degree of the driver's head posture is judged according to the distance between the center of the eyebrows and the chin in the video frame, and the head bowing degree is calculated. The degree of inclination of the driver's head posture is judged according to the change of the angle between the center of the eyebrows and the straight line where the chin is located in the video frame, and the head inclination ratio is calculated.

[0067] After selecting eye feature points, lip feature points, or other feature points, a coordinate axis is established to describe the coordinate information of each feature point.

[0068] Furthermore, in visual algorithms, there are generally 68 facial feature points. This method lists 18 key facial feature points, such as Figure 2 As shown, by calculating the aspect ratio of the driver's eyes and lips, the blinking frequency, eye closing times, eye closing time percentage, yawning frequency, etc. are obtained.

[0069] In one embodiment, the driver's eye aspect ratio calculation formula is specifically:

[0070]

[0071]

[0072]

[0073] Among them, EAR L EAR R are the aspect ratios of the driver's left eye and right eye respectively, EAR is the driver's average eye aspect ratio, and p1, p2, p3, p4, p5, p6, p7, p8, p9, p10, p11, and p12 are the coordinate information of the eye feature points.

[0074] In one embodiment, the driver's lip aspect ratio calculation formula is specifically:

[0075]

[0076] Among them, MAR is the aspect ratio of the driver's lips, and p13, p14, p15, p16, p17, and p18 are the coordinate information of the lip feature points.

[0077] In one embodiment, the eye aspect ratio is set with an eye aspect ratio threshold EARthreshold , based on the comparison between the driver's eye aspect ratio and the eye aspect ratio threshold, the number of eye closures and eye closure time are calculated, and then the eye confidence is calculated. eye ;

[0078] The lip aspect ratio is set with a lip aspect ratio threshold MAR threshold According to the driver's lip aspect ratio and lip aspect ratio threshold, the number of yawns and yawning time are calculated, and then the lip confidence is calculated. mouth .

[0079] In one embodiment, the eye confidence calculation specifically includes: counting each video frame of the driver while driving, and when EAR < EAR threshold When the eye is closed, it is judged as open. The closed eye time is calculated by the number of closed eye times in consecutive video frames, and then the eye confidence in the fatigue monitoring algorithm is calculated by the percentage of closed eye time in the total time. eye ;

[0080] The lip confidence calculation specifically includes: counting each video frame when the driver is driving, and when MAR<MAR threshold When the mouth is closed, it is judged as open. The mouth opening time is calculated by the number of mouth opening times in consecutive video frames, and then the lip confidence in the fatigue monitoring algorithm is calculated by the percentage of mouth opening time in the total time. mouth .

[0081] Specifically, when the facial features also include head facial features, the degree of the driver's head bowing posture is judged according to the distance between the center of the eyebrows and the chin in the video frame, and then the number of head bowing times and the time of head bowing are calculated according to the set head bowing ratio threshold, and then the head confidence in the fatigue monitoring algorithm is calculated by the percentage of head bowing time in the total time. head .

[0082] The tilt degree of the driver's head posture can also be determined based on the angle change between the center of the eyebrows and the straight line where the chin is located in the video frame, and the head tilt ratio can be calculated. Then, according to the set head tilt ratio threshold, the number of head tilts and tilt time are calculated, and then the head confidence in the fatigue monitoring algorithm is calculated by the percentage of head tilt time in the total time. head .

[0083] In one embodiment, the specific algorithm of the fatigue indicator in the fatigue monitoring algorithm is:

[0084] Confidence=a*Confidenceeye +b*Confidence mouth

[0085] Wherein, a and b are the proportion parameters of the set eye confidence and lip confidence in the total confidence, respectively, and a+b=1.

[0086] Furthermore, the indicators for monitoring and calculating by the adaptive detection method for judging fatigue driving include not only the eye confidence and the lip confidence, but can be adjusted according to actual needs. When the calculated indicators include not only the eye confidence and the lip confidence, the specific algorithm of the fatigue indicator in the fatigue monitoring algorithm can be added with other confidences. At this time, the formula becomes:

[0087] Confidence=a*Confidence eye +b*Confidence mouth +c*Confidence othe

[0088] Among them, a, b and c are the parameters of the proportion of the set eye confidence, lip confidence and other confidences in the total confidence, and a+b+c=1.

[0089] Exemplarily, when the driver's head posture is considered, the other confidence level is the head confidence level, the driver's head movement trajectory is monitored in real time, and the head posture confidence level is calculated based on the head changes in consecutive video frames.

[0090] When other driving postures of the driver except eyes and lips are not considered, c can be set to 0.

[0091] In one embodiment, the driver's eye type includes a first eye type and a second eye type, and the driver's eye type is provided with an eye morphology recognition threshold;

[0092] When the eye aspect ratio of the left eye and the right eye of the driver is greater than the set eye morphology recognition threshold when the driver is not fatigued, the driver's eye type belongs to the second eye type; otherwise, it belongs to the first eye type.

[0093] Specifically, when the driver's eye type is in the first eye type, the system will judge the driver as a small-eye type driver, and when the driver's eye type is in the second eye type, the system will judge the driver as a normal type driver.

[0094] In one embodiment, when the driver's eye type belongs to the first eye type, the eye confidence ratio parameter a in the fatigue monitoring algorithm is reduced;

[0095] When the driver belongs to the second eye type, the parameter a of the eye confidence ratio in the fatigue monitoring algorithm is increased.

[0096] For example, when the original eye confidence ratio parameter is 50%, the fatigue driving adaptive detection method finds that the driver is not driving fatigued after receiving the fatigue driving signal, and the driver belongs to the first eye type. At this time, the eye confidence a can be reduced by 10%, and so on, until the eye confidence ratio parameter is consistent with the driver's current status, or until the eye confidence ratio drops to 0 or drops to the set minimum eye confidence threshold.

[0097] Among them, the decreasing speed of eye confidence can be set by yourself, and the decreasing speed can be changed according to specific needs.

[0098] In the description of the present application, it should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. For ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The technology, methods and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, methods and equipment should be regarded as part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0099] It should be noted that, in the present application, the terms "comprises", "includes" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. It should also be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0100] The above embodiments are provided for persons familiar with the art to implement or use the present application. Personnel familiar with the art can make various modifications or changes to the above embodiments without departing from the application concept of the present application. Therefore, the protection scope of the present application is not limited to the above embodiments, but should be the maximum scope of the innovative features mentioned in the claims.

Claims

1. A method for adaptive detection of fatigue driving, characterized in that: The fatigue driving adaptive detection method comprises: Step S101, acquiring facial features of the driver while driving, and calculating the driver's fatigue index according to the driver's facial features and a fatigue monitoring calculation method; Step S102, when the calculated fatigue index exceeds the set fatigue index normal threshold, sending a fatigue driving signal to the vehicle; Step S103, if the driver does not have fatigue driving behavior, re-determine the driver's eye type; Step S104, modifying the parameters in the fatigue monitoring calculation method according to the type of the driver's eyes, and adaptively adjusting the fatigue monitoring calculation method; Step S105, using the adjusted fatigue monitoring calculation method to loop through the above steps until the monitoring is normal.

2. The method for adaptive detection of fatigue driving according to claim 1, characterized in that: Acquiring the driver's facial features while driving specifically includes: A video captured in real time while the driver is driving is obtained, and facial feature points of the driver are extracted from continuous video frames, wherein the facial feature points of the driver include eye feature points and lip feature points, and the eye aspect ratio and the lip aspect ratio are calculated based on the eye feature points and the lip feature points.

3. The method for adaptive detection of fatigue driving according to claim 2, characterized in that: The driver's eye feature points include: the leftmost end of the left eye, the rightmost end of the left eye, the upper end of the left eye, the lower end of the left eye, the leftmost end of the right eye, the rightmost end of the right eye, the upper end of the right eye, and the lower end of the right eye; The driver's lip feature points include: the leftmost end of the lip, the rightmost end of the lip, the upper end of the lip, and the lower end of the lip.

4. The method for adaptive detection of fatigue driving according to claim 3, characterized in that: The specific calculation formula for the driver's eye aspect ratio is: Among them, EAR L EAR R are the left eye aspect ratio and the right eye aspect ratio of the driver, EAR is the average eye aspect ratio of the driver, p1, p2, p3, p4, p5, p6, p7, p8, p9, p 10 、p 11 、p 12 It is the coordinate information of the eye feature points.

5. The method for adaptive detection of fatigue driving according to claim 4, characterized in that: The driver's lip aspect ratio calculation formula is specifically: Where MAR is the driver’s lip aspect ratio, p 13 、p 14 、p 15 、p 16 、p 17 、p 18 It is the coordinate information of the lip feature points.

6. The method for adaptive detection of fatigue driving according to claim 5, characterized in that: The eye aspect ratio is set with an eye aspect ratio threshold EAR threshold , based on the comparison between the driver's eye aspect ratio and the eye aspect ratio threshold, the number of eye closures and eye closure time are calculated, and then the eye confidence is calculated. eye ; The lip aspect ratio is set with a lip aspect ratio threshold MAR threshold According to the driver's lip aspect ratio and lip aspect ratio threshold, the number of yawns and yawning time are calculated, and then the lip confidence is calculated. mouth .

7. The method for adaptively detecting fatigue driving according to claim 6, characterized in that: The eye confidence calculation specifically includes: counting each video frame when the driver is driving, and when the EAR <EAR threshold When the eye is closed, it is judged as open. The closed eye time is calculated by the number of closed eye times in consecutive video frames, and then the eye confidence in the fatigue monitoring algorithm is calculated by the percentage of closed eye time in the total time. eye ; The lip confidence calculation specifically includes: counting each video frame when the driver is driving, and when MAR<MAR threshold When the mouth is closed, it is judged as open. The mouth opening time is calculated by the number of mouth opening times in consecutive video frames, and then the lip confidence in the fatigue monitoring algorithm is calculated by the percentage of mouth opening time in the total time. mouth .

8. The method for adaptive detection of fatigue driving according to claim 6, characterized in that: The specific algorithm of fatigue index in fatigue monitoring algorithm is: Confidence=a*Confidence eye +b*Confidence mouth Wherein, a and b are the proportion parameters of the set eye confidence and lip confidence in the total confidence, respectively, and a+b=1.

9. The method for adaptive detection of fatigue driving according to claim 8, characterized in that: The driver's eye type includes a first eye type and a second eye type, and the driver's eye type is provided with an eye morphology recognition threshold; When the eye aspect ratio of the left eye and the right eye of the driver is greater than the set eye morphology recognition threshold when the driver is not fatigued, the driver's eye type belongs to the second eye type; otherwise, it belongs to the first eye type.

10. The method for adaptive detection of fatigue driving according to claim 9, characterized in that: When the driver's eye type belongs to the first eye type, reducing the eye confidence ratio parameter a in the fatigue monitoring algorithm; When the driver belongs to the second eye type, the parameter a of the eye confidence ratio in the fatigue monitoring algorithm is increased.

Images