Method and system for detecting fatigue driving based on tunnel light

Abstract

The invention belongs to the technical field of tunnel traffic safety, and relates to a method and system for detecting fatigue driving based on tunnel light. Pupil area images of a driver in a non-fatigue state and a real vehicle driving state are obtained, the pupil area images are analyzed and calculated through an image processing algorithm, a pupil area change track is drawn, a reference track and a real vehicle driving track are obtained, the real vehicle driving track is compared with the reference track, and if the comparison result meets the requirement, the driver is judged to be in a non-fatigue state. If yes, the driver is in a normal driving state; if the comparison result does not meet the requirement, the driver is in a fatigue driving state. According to the method, whether the driver is in the fatigue driving state or not is detected through illumination light in the tunnel environment, so that the safety of the driver and passengers in the tunnel is guaranteed, and the method can be applied to different types of tunnels and is wide in application range.

Description

Technical Field

[0001] The invention belongs to the technical field of tunnel traffic safety, and relates to a method and system for detecting fatigue driving based on tunnel lights. Background Art

[0002] Tunnels are closed driving environments, making them more prone to traffic accidents than normal roads. The difference in brightness inside and outside the tunnel may cause drivers to experience "black hole" and "white hole" effects. This sudden change in brightness can cause temporary visual impairment and increase psychological stress on drivers. When drivers enter a tunnel in a fatigued state, their cognitive and operational abilities will not be sufficient to handle the negative impact of driving caused by this sudden change in the environment, which can easily lead to tunnel traffic accidents.

[0003] In addition, when the driver drives in the tunnel for a long time, the poor air circulation in the tunnel may cause the oxygen content in the car to drop. Coupled with the single tunnel environment, these factors will make the driver's brain inactive. Due to the lack of sensory stimulation, the driver will gradually become mentally fatigued, slow to react, and sleepy, causing visual fatigue, inattention, and slow response to emergencies. In severe cases, it may lead to traffic accidents and threaten the lives of drivers and passengers in the tunnel. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and to provide a method and system for detecting fatigue driving based on tunnel lights.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for detecting fatigue driving based on tunnel lights obtains pupil area images of a driver in a non-fatigue state and in an actual vehicle driving state, analyzes and calculates the pupil area images through an image processing algorithm and draws a pupil area change trajectory, obtains a reference trajectory and an actual vehicle driving trajectory, and compares the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigue driving state.

[0007] Further, the following steps are included:

[0008] Step 1: Obtain all pupil area images a of the driver passing through the tunnel in a non-fatigue state, and obtain all pupil area images b of the driver passing through the tunnel in a real driving state;

[0009] Step 2: preprocessing the pupil area image a and the pupil area image b; analyzing and calculating the image in the pupil area image a by an image processing algorithm to obtain a constant value A0 of the driver's pupil area in the pupil area image at time t, and analyzing and calculating the image in the pupil area image b by an image processing algorithm to obtain a pupil area A(t) of the driver in the pupil area image at time t;

[0010] Step 3: Draw the pupil area change trajectory under the non-fatigue driving state through the constant value A0 of the driver's pupil area, and mark the pupil area change trajectory under the non-fatigue driving state as the reference trajectory;

[0011] The pupil area change trajectory of the driver under the actual vehicle driving state is plotted using the driver's pupil area A(t), and the pupil area change trajectory under the actual vehicle driving state is marked as the actual vehicle driving trajectory;

[0012] Step 4: Compare the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigue driving state.

[0013] Furthermore, the pupil area image a and the pupil area image b in the step 1 are both captured by an infrared camera.

[0014] Furthermore, in the step 2, the preprocessing includes filtering, graying, and binarizing the image in sequence.

[0015] Furthermore, in step 2, the image processing algorithm first detects the edge of the pupil using an edge detection algorithm, then uses a contour search method to find the contour of the pupil, and finally calculates the area of ​​the screened pupil contour.

[0016] Furthermore, the area calculation includes calculating the pixel area of ​​each picture:

[0017] Calculate the area P of each pixel:

[0018] P = px × py (1)

[0019] Where: px is the width of each pixel in the image, py is the height of each pixel in the image.

[0020] Furthermore, in step 3, the pupil area A(t) of the driver in the pupil area image at time t is:

[0021] A(t)=A0+f(ΔV(t))+g(d(t)-d0)(4)

[0022] Where: A0 is the constant value of pupil area in the non-fatigue state; ΔV(t) is the rate of change of vehicle speed; f(ΔV(t)) is the function of the change of pupil area with respect to vehicle speed; d(t) is the distance between the vehicle and the tunnel wall at time t; d0 is the normal distance between the vehicle and the wall in the non-fatigue state.

[0023] Furthermore, in step 4, the actual vehicle driving trajectory is compared with the reference trajectory by comparing the maximum value A(t)max and the minimum value A(t)min of the pupil area A(t) in the actual vehicle driving trajectory with a specific multiple of A0, respectively, and the comparison rules are as follows:

[0024]

[0025] Furthermore, the present invention also provides a detection system for the method of detecting fatigue driving based on tunnel lighting, characterized in that it includes:

[0026] Image acquisition module: used to acquire all pupil area images a of the driver passing through the tunnel in a non-fatigue state, and acquire all pupil area images b of the driver passing through the tunnel in a real driving state;

[0027] Analysis and calculation module: used to calculate the constant value A0 of the driver's pupil area in the pupil area image at time t, and the driver's pupil area A(t) in the pupil area image at time t;

[0028] Drawing module: used to draw the reference trajectory and the actual vehicle driving trajectory;

[0029] Comparison module: compares the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigued driving state.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The present invention utilizes the lighting in the tunnel environment to detect whether the driver is in a fatigue driving state, thereby ensuring the safety of the driver and passengers in the tunnel. The method can be used in different types of tunnels and has a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings are incorporated in and constitute a part of this specification and, together with the description, serve to explain the principles of the present invention.

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0034] Figure 1 This is a schematic diagram of the principle of the method for detecting fatigue driving based on tunnel lights of the present invention;

[0035] Figure 2 The figure is a schematic diagram of a specific process of the method for detecting fatigue driving based on tunnel lights of the present invention. DETAILED DESCRIPTION

[0036] Here, exemplary embodiments will be described in detail, and the embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are only examples of devices consistent with some aspects of the present invention as detailed in the appended claims.

[0037] A method for detecting fatigue driving based on tunnel lights obtains pupil area images of a driver in a non-fatigue state and in an actual vehicle driving state, analyzes and calculates the pupil area images through an image processing algorithm and draws a pupil area change trajectory, obtains a reference trajectory and an actual vehicle driving trajectory, and compares the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigue driving state.

[0038] Furthermore, if Figure 2 As shown, the following steps are included:

[0039] Step 1: Obtain all pupil area images a of the driver passing through the tunnel in a non-fatigue state, and obtain all pupil area images b of the driver passing through the tunnel in a real driving state;

[0040] Step 2: preprocessing the pupil area image a and the pupil area image b; analyzing and calculating the image in the pupil area image a by an image processing algorithm to obtain a constant value A0 of the driver's pupil area in the pupil area image at time t, and analyzing and calculating the image in the pupil area image b by an image processing algorithm to obtain a pupil area A(t) of the driver in the pupil area image at time t;

[0041] Step 3: Draw the pupil area change trajectory under the non-fatigue driving state through the constant value A0 of the driver's pupil area, and mark the pupil area change trajectory under the non-fatigue driving state as the reference trajectory;

[0042] The pupil area change trajectory of the driver under the actual vehicle driving state is plotted using the driver's pupil area A(t), and the pupil area change trajectory under the actual vehicle driving state is marked as the actual vehicle driving trajectory;

[0043] Step 4: Compare the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigue driving state.

[0044] Furthermore, the pupil area image a and the pupil area image b in the step 1 are both captured by an infrared camera.

[0045] Furthermore, in the step 2, the preprocessing includes filtering, graying, and binarizing the image in sequence.

[0046] Furthermore, in step 2, the image processing algorithm first detects the edge of the pupil using an edge detection algorithm, then uses a contour search method to find the contour of the pupil, and finally calculates the area of ​​the screened pupil contour.

[0047] Furthermore, the area calculation includes calculating the pixel area of ​​each picture:

[0048] Calculate the area P of each pixel:

[0049] P=px×py (1)

[0050] Where: px is the width of each pixel in the image, py is the height of each pixel in the image.

[0051] Furthermore, in step 3, the pupil area A(t) of the driver in the pupil area image at time t is:

[0052] A(t)=A0+f(ΔV(t))+g(d(t)-d0) (4)

[0053] Where: A0 is the constant value of pupil area in the non-fatigue state; ΔV(t) is the rate of change of vehicle speed; f(ΔV(t)) is the function of the change of pupil area with respect to vehicle speed; d(t) is the distance between the vehicle and the tunnel wall at time t; d0 is the normal distance between the vehicle and the wall in the non-fatigue state.

[0054] Furthermore, in step 4, the actual vehicle driving trajectory is compared with the reference trajectory by comparing the maximum value A(t)max and the minimum value A(t)min of the pupil area A(t) in the actual vehicle driving trajectory with a specific multiple of A0, respectively, and the comparison rules are as follows:

[0055]

[0056] Furthermore, the present invention also provides a detection system for the method of detecting fatigue driving based on tunnel lighting, characterized in that it includes:

[0057] Image acquisition module: used to acquire all pupil area images a of the driver passing through the tunnel in a non-fatigue state, and acquire all pupil area images b of the driver passing through the tunnel in a real driving state;

[0058] Analysis and calculation module: used to calculate the constant value A0 of the driver's pupil area in the pupil area image at time t, and the driver's pupil area A(t) in the pupil area image at time t;

[0059] Drawing module: used to draw the reference trajectory and the actual vehicle driving trajectory;

[0060] Comparison module: compares the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigued driving state.

[0061] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0062] Example

[0063] The present invention provides a method for detecting fatigue driving based on tunnel lighting, the purpose of which is to utilize the lighting facilities arranged in the tunnel to reversely infer whether the driver has fatigue driving behavior by determining whether there is any abnormality in the change pattern of the driver's pupil area.

[0064] Specifically, when a driver is driving in a tunnel, the effect of tunnel lights on the driver's pupil area follows a certain law of change: when the driver enters a darker tunnel from a bright external environment, in order to adapt to the lower light level, the pupil will naturally expand, that is, the pupil area will increase, so that more light can enter the eye and improve vision in a dark environment; as the driver drives in the tunnel, the pupil will gradually adjust to a state of adapting to the current light level. Since the lights in the tunnel are generally arranged according to a certain distance rule, when passing a light, there is a process of approaching the light and then moving away from the light. In this process, the intensity of light received by the driver's eyes first increases and then decreases. At this time, the size of the driver's pupil area will also have a certain law of change, that is, the pupil area will fluctuate with the change of light. When approaching the bright area of ​​the light, the pupil area will shrink, and when moving away from the darker area of ​​the light, the pupil area will increase; when the driver is about to leave the tunnel and enter the bright external environment, the pupil area will decrease rapidly to reduce the amount of light entering the eye and avoid glare and discomfort caused by strong light.

[0065] The above-mentioned change in pupil area is controlled by the autonomic nervous system and is an unconscious physiological reaction. This adjustment function of the pupil is crucial for the driver's visual adaptation in an environment with changing light, but at the same time, frequent light changes and pupil adjustment may also cause visual fatigue. Based on this physiological reaction, the change pattern of the driver's pupil caused by the change of tunnel light can be used to reversely infer whether the driver is experiencing fatigue driving behavior.

[0066] To achieve the above object, the present invention provides the following technical solutions:

[0067] First, through a preliminary experiment, the driver's pupil area change pattern caused by changes in tunnel light intensity when the driver is not fatigued driving is collected, and this pupil area change pattern is used as the judgment benchmark. In the subsequent actual vehicle driving process, when the system detects that the driver's pupil area change amplitude is significantly abnormal compared with the benchmark pattern, it is considered that the driver is fatigued driving behavior.

[0068] like Figure 1 As shown, the specific method is as follows:

[0069] Step 1. The driver conducts a tunnel real-car pre-experimental driving task, and uses an infrared pupil tracker to measure the driver's pupil data when passing through the tunnel in a non-fatigued state. The infrared camera captures high-resolution images of the eyes, which pay special attention to the pupil area. The design and operation of the infrared pupil tracker ensure that the pupil area can be accurately and continuously measured even in low-light or no-light conditions. Before the test begins, the infrared pupil tracker device is calibrated to ensure that it can accurately measure the pupil area;

[0070] Step 2: The images captured by the infrared pupil tracker are analyzed by specific image processing algorithms that can identify the edge of the pupil and use edge detection algorithms or threshold-based segmentation techniques to identify the outline of the pupil. Once the edge of the pupil is detected, the algorithm will draw the outline of the pupil. After the outline of the pupil is identified, the algorithm will calculate the area of ​​the area within the outline to obtain the driver's pupil area;

[0071] Specifically, the captured image is preprocessed, including grayscale, filtering, binarization, etc., to facilitate subsequent contour detection. If the image has large noise and the color information is useless, it can be filtered first and then grayscaled and binarized; if only the shape of the target object needs to be extracted, it can be directly grayscaled and then binarized. It is necessary to select and use them in combination according to the specific situation and image characteristics. Edge detection algorithms such as the Canny algorithm detect the edge of the pupil, and use contour search algorithms such as find Contours to find the contour of the pupil. The correct pupil contour is selected according to the shape and size characteristics of the pupil, and the area of ​​the selected pupil contour is calculated. The following is the formula for calculating the pupil area:

[0072] Assume that A is the actual area of ​​the pupil, N is the number of pixels within the filled outline, px is the width of each pixel in the image, py is the height of each pixel in the image, f is the focal length of the camera lens, and D is the distance from the camera to the pupil.

[0073] First, calculate the actual area P of each pixel:

[0074] P=px×py (1)

[0075] Then, calculate the number of pixels N within the filled contour. Finally, if you need to convert the pixel area to the actual physical area A, you can use the following formula:

[0076]

[0077] Step 3, recording the pupil area data of the entire tunnel, obtaining the pupil area change trajectory under non-fatigue driving conditions, and using it as a reference trajectory for judging whether the driver is fatigued for subsequent analysis;

[0078] Step 4: When the driver drives the vehicle through the tunnel, step 1 and step 2 are repeated to obtain the pupil area change trajectory of the driver, and compare it with the pupil area change reference trajectory in step 3. If there is a large abnormality between the two trajectories, it means that the pupil area no longer follows the normal change law of passing through the tunnel lane, and it is considered that the driver is fatigued. Specifically, whether the driver is fatigued is mainly reflected in whether the vehicle speed is stable and the ability to maintain the lane.

[0079] Step 4.1: When the driver is driving at a constant speed in the tunnel, the lighting conditions are stable and the pupil area will gradually stabilize at a relatively constant level. However, when the driver is tired, the speed control is not stable, that is, the speed is changed in the tunnel, especially when accelerating in a dark environment. The light changes are more sudden and the pupil dilation reaction will be more obvious, causing the pupil area to fluctuate up and down.

[0080] Assume that A(t) represents the pupil area at time t, and V(t) represents the vehicle speed at the same time. When the driver is not tired and the vehicle speed is stable, the pupil area A(t) should be maintained at a relatively constant level A0. When the driver is tired and the vehicle speed is unstable, the fluctuation of the pupil area can be expressed as:

[0081] ΔA v (t) = A(t) - A0(6)

[0082] The light changes caused by unstable vehicle speed can be expressed by the rate of change of vehicle speed:

[0083] ΔV(t)=V(t)-V0(7)

[0084] Where V0 is the speed of the driver when driving at a constant speed. Therefore, the change in pupil area can be considered as a function of the change in vehicle speed:

[0085] ΔA v (t) = f(ΔV(t)) (8)

[0086] Specifically, if the vehicle speed suddenly increases, the pupil area will increase because the light suddenly dims; conversely, when the vehicle speed decreases, the pupil area decreases.

[0087] Step 4.2: When the driver is driving normally, the lane remains unchanged, that is, the change of pupil area under the influence of light is regular. Once the driver is driving fatigued, the lane remains unstable and swings back and forth laterally in the lane. When the vehicle approaches the tunnel wall, due to the angle of light, the driver receives less light and the pupil area will further increase. Conversely, when the vehicle moves away from the wall, the light intensity increases and the pupil area decreases, which will cause the pupil area to fluctuate greatly.

[0088] Assuming d(t) represents the distance between the vehicle and the tunnel wall at time t, and d0 is the normal distance between the vehicle and the wall, the change in pupil area caused by unstable lane keeping can be expressed as:

[0089] ΔA d (t) = A(t) - A0(9)

[0090] Therefore, the change in pupil area can be considered as a function of distance:

[0091] ΔA d (t) = g(d(t) - d0) (10)

[0092] Specifically, if the vehicle swings back and forth laterally in the lane, when the vehicle approaches the wall, the light will decrease and the pupil area will increase; when the vehicle moves away from the wall, the light will increase and the pupil area will decrease.

[0093] Combining the above two factors, a comprehensive model can be obtained to describe the changes in pupil area when the driver is tired:

[0094] A(t)=A0+f(ΔV(t))+g(d(t)-d0) (4)

[0095] In summary, changes in vehicle speed and position in the lane in the tunnel will affect the intensity of light received by the driver, thereby affecting changes in pupil area.

[0096] According to the above model, when the calculated pupil area change A(t) satisfies the following range:

[0097] A(t) max>1.2A0(2)

[0098] or,

[0099] A(t) min <0.8A0(3)

[0100] This means that the driver is changing speed in the tunnel, the lane is unstable, and the vehicle is swinging back and forth laterally in the lane. At this time, it is considered that the driver is currently in a fatigue driving state.

[0101] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention.

[0102] It should be understood that the present invention is not limited to what has been described above and that various modifications and changes may be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for detecting fatigue driving based on tunnel lights, characterized in that: The pupil area images of the driver in the non-fatigue state and the actual vehicle driving state are obtained, and the pupil area images are analyzed and calculated through the image processing algorithm to draw the pupil area change trajectory, and the baseline trajectory and the actual vehicle driving trajectory are obtained. The actual vehicle driving trajectory is compared with the baseline trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigued driving state.

2. The method for detecting fatigue driving based on tunnel lights according to claim 1 is characterized in that: The following steps are involved: Step 1: Obtain all pupil area images a of the driver passing through the tunnel in a non-fatigue state, and obtain all pupil area images b of the driver passing through the tunnel in a real driving state; Step 2: preprocessing the pupil area image a and the pupil area image b; The image in the pupil area image a is analyzed and calculated by an image processing algorithm to obtain a constant value A0 of the driver's pupil area in the pupil area image at time t; the image in the pupil area image b is analyzed and calculated by an image processing algorithm to obtain a pupil area A(t) of the driver in the pupil area image at time t; Step 3: Draw the pupil area change trajectory under the non-fatigue driving state through the constant value A0 of the driver's pupil area, and mark the pupil area change trajectory under the non-fatigue driving state as the reference trajectory; The pupil area change trajectory of the driver under the actual vehicle driving state is plotted through the driver's pupil area A(t), and the pupil area change trajectory under the actual vehicle driving state is marked as the actual vehicle driving trajectory; Step 4: Compare the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigue driving state.

3. The method for detecting fatigue driving based on tunnel lights according to claim 2 is characterized in that: The pupil area image a and the pupil area image b in the step 1 are both captured by an infrared camera.

4. The method for detecting fatigue driving based on tunnel lights according to claim 2 is characterized in that: In the step 2, the preprocessing includes filtering, graying, and binarizing the image in sequence.

5. The method for detecting fatigue driving based on tunnel lights according to claim 2 is characterized in that: In step 2, the image processing algorithm first detects the edge of the pupil using an edge detection algorithm, then uses a contour search method to find the contour of the pupil, and finally calculates the area of ​​the selected pupil contour.

6. The method for detecting fatigue driving based on tunnel lights according to claim 5 is characterized in that: The area calculation includes calculating the pixel area of ​​each image: Calculate the area P of each pixel: P = px × py (1) Where: px is the width of each pixel in the image, py is the height of each pixel in the image.

7. The method for detecting fatigue driving based on tunnel lights according to claim 2 is characterized in that: In step 3, the pupil area A(t) of the driver in the pupil area image at time t is: A(t)=A0+f(ΔV(t))+g(d(t)-d0)(4) Where: A0 is the constant value of pupil area in the non-fatigue state; ΔV(t) is the rate of change of vehicle speed; f(ΔV(t)) is the function of the change of pupil area with respect to vehicle speed; d(t) is the distance between the vehicle and the tunnel wall at time t; d0 is the normal distance between the vehicle and the wall in the non-fatigue state.

8. The method for detecting fatigue driving based on tunnel lights according to claim 2 is characterized in that: In the step 4, the actual vehicle driving trajectory is compared with the reference trajectory by comparing the maximum value A(t)max and the minimum value A(t)min of the pupil area A(t) in the actual vehicle driving trajectory with a specific multiple of A0, respectively. The comparison rules are as follows:

9. A detection system based on the method for detecting fatigue driving based on tunnel lights according to any one of claims 1 to 8, characterized in that: include: Image acquisition module: used to acquire all pupil area images a of the driver passing through the tunnel in a non-fatigue state, and acquire all pupil area images b of the driver passing through the tunnel in a real driving state; Analysis and calculation module: used to calculate the constant value A0 of the driver's pupil area in the pupil area image at time t, and the driver's pupil area A(t) in the pupil area image at time t; Drawing module: used to draw the reference trajectory and the actual vehicle driving trajectory; Comparison module: compares the actual vehicle driving trajectory with the reference trajectory. If the comparison result meets the requirements, the driver is in a normal driving state; if the comparison result does not meet the requirements, the driver is in a fatigued driving state.

Images