Optimized driving fatigue monitoring method and device

Abstract

The invention discloses an optimized driving fatigue monitoring method, which introduces face position prediction on the basis of performing feature extraction on a driving image by adopting face detection, and determines whether to perform feature extraction on the driving image by adopting face detection according to a prediction confidence coefficient of a face prediction position. According to the method, the driving image is subjected to feature extraction through face prediction or face prediction, and only when the prediction confidence of the face prediction position does not reach the confidence threshold, the driving image is subjected to feature extraction through face prediction, so that the number of times of feature extraction of the driving image through face prediction is reduced, and the accuracy of feature extraction is improved. And the computing power and the time are saved. The invention also discloses a device for realizing the driving fatigue monitoring method.

Description

Technical Field

[0001] The present invention relates to safe driving technology, and in particular to an optimized driving fatigue monitoring method and device. Background Art

[0002] Fatigue driving is one of the important causes of traffic accidents. When a driver is in a fatigued state, his attention will be distracted, his ability to respond will decrease, he will not observe the road conditions and surrounding environment in a timely manner, and he may even fall asleep for a short time and completely lose his driving ability, which can easily cause traffic accidents.

[0003] Among the "risk factors affecting traffic accidents" counted in the World Report on Road Traffic Injury Prevention, driving too fast or at an inappropriate speed, driving under the influence of alcohol, and driving while fatigued are the main causes of traffic accidents, deaths, and serious injuries to all vehicle occupants. According to statistics from my country's transportation department, traffic accidents caused by fatigue driving account for about 20% of total traffic accidents, 40% of major traffic accidents, and 83% of traffic deaths.

[0004] The China New Car Assessment Program (C-NCAP Management Rules (2024 Edition)) will be officially implemented on July 1, 2024, which will add a driver monitoring system (DMS) assessment item to the active safety section. Article 6 of the European Union (EU) Regulation 2019 / 2144 requires that M and N motor vehicles must be equipped with advanced vehicle systems, including a driver drowsiness and attention warning (DDAW) system. The DDAW system should be able to monitor the driver's drowsiness and warn the driver through the HMI (human-machine interface). The DDAW system's privacy and data protection processes should comply with the requirements of the EU Data Protection Law GDPR. The DDAW regulation was officially promulgated and took effect in August 2021. The regulation will be mandatory for all new models from July 2022 and for all new vehicles from July 2024.

[0005] The vision-based DMS system adopts direct monitoring and non-contact solutions to characterize the driver's fatigue level by collecting image signals such as the driver's face, eyes, and head movements. Compared with indirect monitoring based on vehicle behavior characteristics, direct monitoring is more accurate and real-time, and can detect fatigue status faster. For example, the Chinese patent with application number CN202410210718.7 discloses a driving fatigue detection method based on lightweight neural network image enhancement, including: real-time acquisition of dynamic images of the driver in the cockpit, and enhancement of the dynamic images; capturing the driver's facial image stream in the enhanced dynamic image through a lightweight fast face detector; inputting the facial image stream into a preset facial key feature point extraction network model to extract the facial key feature points of each frame of the facial image stream, wherein the facial key feature point extraction network model is obtained by constructing and training a heavily parameterized lightweight convolutional network; judging whether the driver has facial fatigue features based on the facial key feature points, and determining that the driver is driving fatigued when the facial fatigue features are detected.

[0006] However, face detection requires processing of the entire dynamic image. The whole process consumes a lot of computing resources and takes a long time to calculate, which seriously affects the timeliness and safety of driving fatigue monitoring. Summary of the invention

[0007] In order to solve the above-mentioned deficiencies of the prior art, the present invention provides an optimized driving fatigue monitoring method and device, which can reduce the number of face detections to save computing power and time.

[0008] The technical problem to be solved by the present invention is achieved through the following technical solutions: An optimized driving fatigue monitoring method comprises the following steps: Step 100: Obtain the Nth driving image, N≥1; Step 200: Determine whether there is a predicted face position in the Nth driving image. If so, proceed to Step 300 to Step 700. If not, proceed to Step 500 to Step 700, wherein the predicted face position in the Nth driving image is obtained based on the prediction of the N-1th driving image. Step 300: extracting features from the Nth driving image according to the predicted face position to obtain first feature information; Step 400: Calculate the prediction confidence according to the first feature information, and determine whether the prediction confidence reaches a confidence threshold. If it does not reach the confidence threshold, proceed to steps 500 to 700. If it reaches the confidence threshold, use the first feature information as facial feature information and proceed to step 700. Step 500: performing face detection on the Nth driving image to obtain a face detection position; Step 600: extracting features from the Nth driving image according to the face detection position to obtain second feature information, and using the second feature information as face feature information; Step 700: Determine whether the driver is driving fatigued based on the facial feature information.

[0009] 1. An optimized driving fatigue monitoring method, characterized in that it comprises the following steps: Step 100: Obtain the Nth driving image, N≥1; Step 200: Determine whether there is a predicted face position in the Nth driving image. If so, proceed to Step 300 to Step 700. If not, proceed to Step 500 to Step 700, wherein the predicted face position in the Nth driving image is obtained based on the prediction of the N-1th driving image. Step 300: extracting features from the Nth driving image according to the predicted face position to obtain first feature information; Step 400: Calculate prediction confidence based on the first feature information, and determine whether the prediction confidence reaches a confidence threshold. If it does not reach the confidence threshold, proceed to steps 500 to 700. If it reaches the confidence threshold, use the first feature information as facial feature information and proceed to step 700. Step 500: performing face detection on the Nth driving image to obtain a face detection position; Step 600: extracting features from the Nth driving image according to the face detection position to obtain second feature information, and using the second feature information as face feature information; Step 700: Determine whether the driver is driving fatigued based on the facial feature information.

[0010] Furthermore, in step 300, the steps of extracting features from the Nth driving image according to the predicted face position are as follows: Step 310: performing face cropping on the Nth driving image according to the predicted face position to obtain a first face image; Step 320: Perform feature extraction on the first facial image to obtain the first feature information.

[0011] Furthermore, in step 310, when performing face cropping on the Nth driving image, a corresponding face recognition frame is set in the Nth driving image to crop the image based on the predicted coordinate information of the feature point in the Nth image and the length and width of the face in the N-1th driving image.

[0012] Furthermore, the predicted face position includes coordinate information of at least one feature point in the Nth driving image, and the coordinate information is calculated using a position prediction model based on coordinate information of the same feature point in the N-1th driving image and the time difference between the two driving images.

[0013] Furthermore, in step 600, the steps of extracting features from the Nth driving image according to the face detection position are as follows: Step 610: performing face cropping on the Nth driving image according to the face detection position to obtain a second face image; Step 620: Perform feature extraction on the second facial image to obtain the second feature information.

[0014] Furthermore, a feature extraction network is used to extract features from the first face image and the second face image. The backbone network of the feature extraction network adopts an improved mobilenet network and is combined with a DenseNet network. At the same time, some modules are changed into lighter ghost modules, and the accuracy is improved by adding convolutional layers.

[0015] Furthermore, in step 700, the driver's fatigue level is calculated based on the proportion of time that the eyes are closed per unit time, the facial feature information includes eyelid edge feature points and pupil edge feature points, and the proportion of the pupil area covered by the eyelids is calculated based on the coordinate information of the eyelid edge feature points and the pupil edge feature points. When the area proportion reaches a first proportion threshold, it is judged that the eyes are in a closed state, and the proportion of time that the eyes are closed per unit time is counted. When the time proportion reaches a second proportion threshold, it is judged that the driver is driving fatigued.

[0016] Furthermore, after step 700, the following steps are also included: Step 800: Based on the facial feature information, perform position prediction on the N+1th driving image to obtain the predicted face position of the N+1th driving image.

[0017] An optimized driving fatigue monitoring device is used to implement the above-mentioned driving fatigue monitoring method; the driving fatigue monitoring device comprises: An image acquisition module, used for acquiring driving images; A position prediction module, used to perform face prediction on the next driving image based on the previous driving image to obtain a predicted face position; A position detection module, used for performing face detection on the driving image to obtain a face detection position; a feature extraction module, configured to extract features of the driving image according to the predicted face position or the detected face position, so as to obtain first feature information or second feature information; a confidence module, configured to determine, based on the prediction confidence of the first feature information, the first feature information or the second feature information as facial feature information; The fatigue judgment module is used to judge whether the driver is driving fatigued according to the facial feature information.

[0018] An optimized driving fatigue monitoring device comprises a processor and a memory connected to the processor, wherein the memory stores a computer program for execution by the processor, and when the processor executes the computer program, the above-mentioned driving fatigue monitoring method is performed.

[0019] The present invention has the following beneficial effects: the driving fatigue monitoring method of the present invention introduces face position prediction on the basis of using face detection to extract features from driving images, and determines whether to use face detection or face prediction to extract features from the driving images according to the prediction confidence of the face prediction position. Only when the prediction confidence of the face prediction position does not reach the confidence threshold, face prediction is used to extract features from the driving images, thereby reducing the number of times face prediction is used to extract features from the driving images, thereby saving computing power and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of the driving fatigue monitoring method provided by the present invention.

[0021] Figure 2 This is a schematic diagram of the architecture of the feature extraction network provided by the present invention.

[0022] Figure 3 This is a schematic diagram of the architecture of the separable convolutional layer in the feature extraction network provided by the present invention.

[0023] Figure 4 This is a functional block diagram of the driving fatigue monitoring device provided by the present invention.

[0024] Figure 5 This is a functional block diagram of another driving fatigue monitoring device provided by the present invention. DETAILED DESCRIPTION

[0025] The present invention is described in detail below in conjunction with the accompanying drawings and embodiments, examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0026] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0027] In addition, the terms "first", "second", and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first", "second", and "third" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0028] In the present invention, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", "fix", "set" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0029] Embodiment 1 like Figure 1 As shown, an optimized driving fatigue monitoring method comprises the following steps: Step 100: Obtain the Nth driving image, where N≥1.

[0030] In step 100, a monitoring camera may be arranged in front of the cab of the vehicle (such as at the A-pillar position of the vehicle, the upper position of the front windshield, etc.), so as to use the monitoring camera to take real-time photos of the driver while the vehicle is driving, so as to obtain the driving image.

[0031] Preferably, the surveillance camera is an infrared camera.

[0032] Step 200: Determine whether there is a face prediction position in the Nth driving image. If so, proceed to steps 300 to 700. If not, proceed to steps 500 to 700, wherein the face prediction position in the Nth driving image is obtained based on the prediction of the N-1th driving image.

[0033] In step 200, for the first driving image, since there is no zeroth driving image, it is naturally impossible to predict the face position of the first driving image based on the zeroth driving image, that is, there is no predicted face position of the first driving image, so steps 300 to 700 are performed to calculate the fatigue level of the first driving image through face detection; for the second and subsequent driving images, since there are the first and subsequent driving images, it is naturally possible to predict the face positions of the second and subsequent driving images respectively based on the first and subsequent driving images, so steps 500 to 700 are performed to calculate the fatigue level of the second and subsequent driving images through face prediction or face detection.

[0034] Step 300: extract features of the Nth driving image according to the predicted face position to obtain first feature information.

[0035] In step 300, the predicted face position includes coordinate information of at least one feature point in the Nth driving image, and the coordinate information is calculated using a position prediction model based on coordinate information of the same feature point in the N-1th driving image and the time difference between the two driving images.

[0036] In this embodiment, the feature points used for position prediction include but are not limited to the intersection point between the central axis of the face and the line connecting the two eyes; the position prediction model can be but is not limited to a Kalman filter model, a Markov chain model or a particle filter model.

[0037] Among them, in step 300, the steps of extracting features from the Nth driving image according to the predicted face position are as follows: Step 310: Perform face cropping on the Nth driving image according to the predicted face position to obtain a first face image.

[0038] In step 310, since the predicted face position is obtained based on the prediction of the N-1th driving image, when feature extraction is performed on the N-1th driving image, the extracted face feature information includes the coordinate information of the feature points and the length and width of the face. When face cropping is performed on the Nth driving image, a corresponding face recognition frame can be set in the Nth driving image to crop the image based on the predicted coordinate information of the feature points in the Nth image and the length and width of the face in the N-1th driving image.

[0039] Step 320: Perform feature extraction on the first facial image to obtain the first feature information.

[0040] In step 320, a feature extraction network may be used to perform feature extraction on the first face image, and a feature extraction network may be used to perform feature extraction on the second face image. The backbone network of the feature extraction network adopts an improved mobilenet network and combines it with DenseNet. At the same time, some modules are changed into lighter ghost modules, and the accuracy is improved by adding convolutional layers.

[0041] Specifically, Figure 2 and 3 As shown, the mobilenet network includes an input layer, multiple separable convolutional layers, a classification layer and an output layer, wherein the input layer, multiple separable convolutional layers, the classification layer and the output layer are connected in sequence, wherein the output end of each separable convolutional layer is connected to the classification layer in addition to being connected to the input end of the next separable convolutional layer; each separable convolutional layer includes multiple ghost modules, and the input end of each ghost module is connected to the output ends of all ghost modules in its upper layer.

[0042] Step 400: Calculate the prediction confidence based on the first feature information, and determine whether the prediction confidence reaches a confidence threshold. If it does not reach the confidence threshold, proceed to steps 500-700. If it reaches the confidence threshold, use the first feature information as facial feature information and proceed to step 700.

[0043] In step 400, the prediction confidence of the predicted face position is judged to determine whether there is a large deviation in the predicted face position obtained based on the prediction of the N-1th driving image. If there is a large deviation, that is, the confidence threshold is not reached, steps 500 to 700 are performed, and the fatigue degree of the Nth driving image is still calculated by face detection. If there is a small deviation, that is, the confidence threshold is reached, step 700 is performed to calculate the fatigue degree of the Nth driving image by face prediction.

[0044] The prediction confidence may be the Euclidean distance between a feature point in the face prediction position of the Nth driving image and the face recognition frame. Of course, the face recognition frame may also be calculated using other parameters according to actual needs, and should not be limited thereto.

[0045] In this embodiment, the feature points used to calculate the prediction confidence include but are not limited to the intersection point between the central axis of the face and the line connecting the two eyes.

[0046] Step 500: Perform face detection on the Nth driving image to obtain a face detection position.

[0047] In step 500, a face detection network may be used to perform face detection on the Nth driving image. The face detection network traverses all pixel points in the Nth driving image to obtain coordinate information, a face recognition frame, and the length and width of the face of all pixel points that constitute the face in the Nth driving image. The entire process requires a large amount of computing power resources and takes a long time to calculate.

[0048] In this embodiment, the face detection network can be but is not limited to a yolo network, a CNN network, a MTCNN network or a RNN network.

[0049] Step 600: extract features of the Nth driving image according to the face detection position to obtain second feature information, and use the second feature information as face feature information.

[0050] In step 600, the face detection position includes coordinate information of all pixels constituting the face, the face recognition frame, and the length and width of the face, which are obtained by traversing all pixels in the Nth driving image through the face detection network.

[0051] In this embodiment, the facial feature information can be but is not limited to using the intersection point between the central axis of the face and the line connecting the two eyes as the feature point; the position prediction model can be but is not limited to a Kalman filter model, a Markov chain model or a particle filter model, etc.

[0052] In step 600, the steps of extracting features from the Nth driving image according to the face detection position are as follows: Step 610: Perform face cropping on the Nth driving image according to the face detection position to obtain a second face image.

[0053] In step 610, the face detection position includes the coordinate information of all feature points and the length and width of the face. When the face is cropped on the Nth driving image, a corresponding face recognition frame can be set in the Nth driving image to crop the image based on the coordinate information of all feature points detected in the Nth image and the length and width of the face.

[0054] Step 620: Perform feature extraction on the second facial image to obtain the second feature information.

[0055] In step 620, a feature extraction network may be used to extract features from the second facial image. The backbone network of the feature extraction network adopts an improved mobilenet network combined with DenseNet. At the same time, some modules are changed into lighter ghost modules, and the accuracy is improved by adding convolutional layers.

[0056] Specifically, Figure 2 and 3 As shown, the mobilenet network includes an input layer, multiple separable convolutional layers, a classification layer and an output layer, wherein the input layer, multiple separable convolutional layers, the classification layer and the output layer are connected in sequence, wherein the output end of each separable convolutional layer is connected to the classification layer in addition to being connected to the input end of the next separable convolutional layer; each separable convolutional layer includes multiple ghost modules, and the input end of each ghost module is connected to the output ends of all ghost modules in its upper layer.

[0057] Step 700: Determine whether the driver is driving fatigued based on the facial feature information.

[0058] In step 700, the driver's fatigue level is calculated based on the proportion of time that the eyes are closed per unit time, the facial feature information includes eyelid edge feature points and pupil edge feature points, and the proportion of the pupil area covered by the eyelids is calculated based on the coordinate information of the eyelid edge feature points and the pupil edge feature points. When the area proportion reaches a first proportion threshold, it is determined that the eyes are in a closed state, and the proportion of time that the eyes are closed per unit time is counted. When the time proportion reaches a second proportion threshold, it is determined that the driver is driving fatigued.

[0059] Step 800: Based on the facial feature information, perform position prediction on the N+1th driving image to obtain the predicted face position of the N+1th driving image.

[0060] In step 800, the facial feature information includes coordinate information of at least one feature point in the Nth driving image, and a position prediction model is used to predict the coordinate information of a certain feature point in the Nth driving image and the time difference between the Nth and N+1th driving images to obtain the coordinate information of the same feature point in the N+1th driving image.

[0061] In this embodiment, the feature points used for position prediction include but are not limited to the intersection point between the central axis of the face and the line connecting the two eyes; the position prediction model can be but is not limited to a Kalman filter model, a Markov chain model or a particle filter model.

[0062] The driving fatigue monitoring method of the present invention introduces face position prediction on the basis of using face detection to extract features from driving images, and determines whether to use face detection or face prediction to extract features from the driving images according to the prediction confidence of the face prediction position. Only when the prediction confidence of the face prediction position does not reach the confidence threshold, face prediction is used to extract features from the driving images, thereby reducing the number of times face prediction is used to extract features from the driving images, thereby saving computing power and time.

[0063] Embodiment 2 like Figure 4 As shown, an optimized driving fatigue monitoring device is used to implement the driving fatigue monitoring method described in Example 1; the driving fatigue monitoring device includes: An image acquisition module, used for acquiring driving images; A position prediction module, used to perform face prediction on the next driving image based on the previous driving image to obtain a predicted face position; A position detection module, used for performing face detection on the driving image to obtain a face detection position; a feature extraction module, configured to extract features of the driving image according to the predicted face position or the detected face position, so as to obtain first feature information or second feature information; a confidence module, configured to determine, based on the prediction confidence of the first feature information, the first feature information or the second feature information as facial feature information; The fatigue judgment module is used to judge whether the driver is driving fatigued according to the facial feature information.

[0064] Embodiment 3 like Figure 5 As shown, an optimized driving fatigue monitoring device includes a processor and a memory connected to the processor, wherein the memory stores a computer program for execution by the processor, and when the processor executes the computer program, the driving fatigue monitoring method described in Example 1 is performed.

[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention rather than to limit them. Although the embodiments of the present invention have been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the embodiments of the present invention can still be modified or replaced by equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An optimized driving fatigue monitoring method, characterized in that: The steps include: Step 100: Obtain the Nth driving image, N≥1; Step 200: Determine whether there is a predicted face position in the Nth driving image. If so, proceed to Step 300 to Step 700. If not, proceed to Step 500 to Step 700, wherein the predicted face position in the Nth driving image is obtained based on the prediction of the N-1th driving image. Step 300: extracting features from the Nth driving image according to the predicted face position to obtain first feature information; Step 400: Calculate prediction confidence based on the first feature information, and determine whether the prediction confidence reaches a confidence threshold. If it does not reach the confidence threshold, proceed to steps 500 to 700. If it reaches the confidence threshold, use the first feature information as facial feature information and proceed to step 700. Step 500: performing face detection on the Nth driving image to obtain a face detection position; Step 600: extracting features from the Nth driving image according to the face detection position to obtain second feature information, and using the second feature information as face feature information; Step 700: Determine whether the driver is driving fatigued based on the facial feature information.

2. The driving fatigue monitoring method according to claim 1, characterized in that: In step 300, the steps of extracting features from the Nth driving image according to the predicted face position are as follows: Step 310: performing face cropping on the Nth driving image according to the predicted face position to obtain a first face image; Step 320: Perform feature extraction on the first facial image to obtain the first feature information.

3. The driving fatigue monitoring method according to claim 2, characterized in that: In step 310, when cropping the face of the Nth driving image, a corresponding face recognition frame is set in the Nth driving image to crop the image according to the predicted coordinate information of the feature point in the Nth image and the length and width of the face in the N-1th driving image.

4. The driving fatigue monitoring method according to any one of claims 1 to 3, characterized in that: The predicted face position includes coordinate information of at least one feature point in the Nth driving image, and the coordinate information is calculated using a position prediction model based on coordinate information of the same feature point in the N-1th driving image and a time difference between two driving images.

5. The driving fatigue monitoring method according to claim 1, characterized in that: In step 600, the steps of extracting features from the Nth driving image according to the face detection position are as follows: Step 610: performing face cropping on the Nth driving image according to the face detection position to obtain a second face image; Step 620: Perform feature extraction on the second facial image to obtain the second feature information.

6. The driving fatigue monitoring method according to claim 1, characterized in that: A feature extraction network is used to extract features from the first face image and the second face image. The backbone network of the feature extraction network adopts an improved mobilenet network and is combined with a DenseNet network. At the same time, some modules are changed into lighter ghost modules, and the accuracy is improved by adding convolutional layers.

7. The driving fatigue monitoring method according to claim 1, characterized in that: In step 700, the driver's fatigue level is calculated based on the proportion of time that the eyes are closed per unit time, the facial feature information includes eyelid edge feature points and pupil edge feature points, and the proportion of the pupil area covered by the eyelids is calculated based on the coordinate information of the eyelid edge feature points and the pupil edge feature points. When the area proportion reaches a first proportion threshold, it is determined that the eyes are in a closed state, and the proportion of time that the eyes are closed per unit time is counted. When the time proportion reaches a second proportion threshold, it is determined that the driver is driving fatigued.

8. The driving fatigue monitoring method according to claim 1, characterized in that: After step 700, the following steps are also included: Step 800: Based on the facial feature information, perform position prediction on the N+1th driving image to obtain the predicted face position of the N+1th driving image.

9. An optimized driving fatigue monitoring device, characterized in that: Used to implement the driving fatigue monitoring method described in claim 1; The driving fatigue monitoring device comprises: An image acquisition module, used for acquiring driving images; A position prediction module, used to perform face prediction on the next driving image based on the previous driving image to obtain a predicted face position; A position detection module, used for performing face detection on the driving image to obtain a face detection position; a feature extraction module, configured to extract features of the driving image according to the predicted face position or the detected face position, so as to obtain first feature information or second feature information; a confidence module, configured to determine, based on the prediction confidence of the first feature information, the first feature information or the second feature information as facial feature information; The fatigue judgment module is used to judge whether the driver is driving fatigued according to the facial feature information.

10. An optimized driving fatigue monitoring device, characterized in that: It comprises a processor and a memory connected to the processor, wherein the memory stores a computer program for execution by the processor, and when the processor executes the computer program, the driving fatigue monitoring method according to claim 1 is performed.

Images