Method for detecting false alarms of an image processing device for a camera

Abstract

The present invention relates to a method for detecting false alarms of an image processing device of a camera (2) located on a vehicle (1), and the image processing device transmits data for modeling marking lines (11, 12). The method for detecting false alarms includes the following steps: · - a first step (101) of determining a first distance (Yi) from the vehicle (1) to the marking line (11, 12) at a first moment (T1), · - a second step (102) of determining a second distance (Y'i) from the vehicle (1) to the marking line (11, 12) at a second moment (T2), · - a step (103) of calculating an initial discontinuity, in which the initial discontinuity is equal to the absolute value of the difference between the first distance (Yi) and the second distance (Y'i), · - a step (104) of detecting an initial discontinuity false alarm, in which the initial discontinuity is compared with a predefined low initial threshold.

Description

Technical Field

[0001] The present invention relates to the field of vehicles equipped with driver assistance systems.

[0002] The present invention relates to a method for detecting false alarms of an image processing device of a camera. Background Art

[0003] Vehicles include at least one camera, and the camera includes an image processing device capable of transmitting modeling data of the marking lines of the vehicle's lane, which is known in the prior art.

[0004] The image processing devices of prior art cameras are not always very reliable, especially when the vehicle is driving on a poorly maintained lane or when there is ambiguity in detecting the markings. False marking line detection may lead to untimely vehicle trajectory correction. This reduces the comfort of vehicle passengers and may cause users to distrust autonomous or semi-autonomous driving systems. Summary of the Invention

[0005] The object of the present invention is to provide a method for detecting false alarms of an image processing device of a camera, so that false alarm detection and associated discomfort can be avoided.

[0006] The present invention relates to a method for detecting false alarms of an image processing device of a camera located on a vehicle 1, the image processing device being capable of transmitting modeling data of the marking lines of the vehicle's lane in a reference system bound to the vehicle, and the method for detecting false alarms includes the following steps:

[0007] - A first step of determining a first distance from the vehicle to the marking line at a first moment according to the modeling data of the marking line,

[0008] - A second step of determining a second distance from the vehicle to the marking line at a second moment according to the modeling data of the marking line,

[0009] - An initial discontinuity calculation step of calculating an initial discontinuity, the initial discontinuity being equal to the absolute value of the difference between the first distance and the second distance,

[0010] - A step of detecting an initial discontinuity false alarm, in which the initial discontinuity is compared with a preset low initial threshold, and if the initial discontinuity is greater than the low initial threshold, an initial discontinuity false alarm is detected.

[0011] According to an aspect of the present invention, the low initial threshold is equal to a value included between 0.1 meter and 0.3 meter, the time difference between the first moment and the second moment is less than or equal to the quotient of the low initial threshold divided by a lateral speed substantially equal to two meters per second, and the time difference is greater than one millisecond.

[0012] According to one aspect of the present invention, the initial discontinuity is also compared with a preset high initial threshold. If the initial discontinuity is greater than the low initial threshold and less than the high initial threshold, an initial discontinuity false alarm is detected.

[0013] According to one aspect of the present invention, the high initial threshold is equal to a value included between 1.7 meters and 2.3 meters.

[0014] According to one aspect of the present invention, the reference system includes a longitudinal coordinate axis substantially parallel to the lateral direction of the vehicle, and an origin flush with the camera of the vehicle. The first distance and the second distance respectively correspond to the ordinates of the points of the marking line at the origin at the first moment and the second moment.

[0015] According to one aspect of the present invention, the reference system includes a transverse coordinate axis substantially parallel to the longitudinal direction of the vehicle. The method for detecting false alarms further includes the following steps:

[0016] - A third step of determining a first ordinate of the horizon point of the marking line according to the modeling data of the marking line. The first abscissa of the horizon point at the first moment is a distance corresponding to the distance that the vehicle will travel within a predefined horizon time.

[0017] - A fourth step of determining a second ordinate of the horizon point of the marking line according to the modeling data of the marking line. The second abscissa of the horizon point at the second moment is a distance corresponding to the distance that the vehicle will travel within the predefined horizon time.

[0018] - A horizon discontinuity calculation step of calculating the horizon discontinuity, where the horizon discontinuity is equal to the absolute value of the difference between the first ordinate and the second ordinate minus the initial discontinuity.

[0019] - A step of detecting a horizon discontinuity false alarm, in which the horizon discontinuity is compared with a preset horizon threshold. If the horizon discontinuity is greater than the horizon threshold, a horizon discontinuity false alarm is detected.

[0020] According to one aspect of the present invention, the predefined horizon time is a value included between 0.9 seconds and 1.1 seconds, and the horizon threshold is a value included between 0.2 meters and 1 meter.

[0021] According to one aspect of the present invention, the modeling data includes the best visible distance. The method for detecting false alarms further includes a step of detecting a range false alarm, in which the best visible distance is compared with a range threshold equal to the distance that the vehicle will travel within a predefined range time. If the best visible distance is less than the range threshold, a range false alarm is detected.

[0022] According to one aspect of the present invention, the predefined range of time is a value included between 0.8 seconds and 0.99 seconds.

[0023] According to one aspect of the present invention, the method for detecting false alarms further includes the following steps: if a false alarm among an initial discontinuity false alarm, a horizon discontinuity false alarm, and a range false alarm is detected, a visual warning is displayed to a user of the vehicle to warn the user that the marking line cannot be detected by the camera temporarily.

[0024] According to one aspect of the present invention, the modeling data of the marking line is of polynomial type, and the degree of the polynomial is greater than or equal to three.

[0025] The present invention also relates to a computer program product including program instructions, which, when executed by a computer, implement at least one step of the method for detecting false alarms.

[0026] The present invention also relates to a readable data medium on which the computer program product is stored. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other advantages and features of the present invention will become apparent by reading the specification and the drawings.

[0028] Figure 1 is a schematic diagram showing a vehicle equipped with an image processing device with a camera traveling on a lane at a first moment.

[0029] Figure 2 is a schematic diagram showing a vehicle equipped with an image processing device with a camera traveling on a lane at a second moment.

[0030] Figure 3 is a schematic diagram showing Figure 1 an alternative embodiment of

[0031] Figure 4 is a schematic diagram showing Figure 2 an alternative embodiment of

[0032] Figure 5 is a schematic diagram showing Figure 2 a variant of

[0033] Figure 6 is a camera image.

[0034] Figure 7 is a schematic diagram showing steps of a method for detecting false alarms of an image processing device of a camera according to the present invention.

[0035] Figure 8a ​​​​​​​​Shows a display screen on which the marking lines of the vehicle's lane are shown in the case of a false alarm where no lane is detected.

[0036] Figure 8b Shows a display screen on which the marking lines of the vehicle's lane are shown in the case of a false alarm where a lane has been detected. Detailed implementation

[0037] Figure 1 Shows a vehicle 1 traveling in lane 10 at a first moment T1. This lane includes two marking lines 11, 12 located on both sides of the vehicle 1.

[0038] Vehicle 1 includes a camera 2, which includes an image processing device capable of transmitting the modeling data of the marking lines 11, 12.

[0039] The modeling data allows the marking lines 11, 12 to be modeled in a reference system 3 bound to the vehicle 1.

[0040] Regardless of whether the markings (e.g., reflective strips) along the marking lines 11, 12 on the ground are continuous, the modeling data allows the marking lines 11, 12 to be modeled as continuous curves.

[0041] Preferably, the modeling data is of polynomial type, and the degree of the polynomial is greater than or equal to three. For example, the image processing device can transmit the coefficients of the polynomial.

[0042] The bound reference system 3 includes a longitudinal coordinate axis Y that is substantially parallel to the lateral direction of the vehicle 1 and a transverse coordinate axis X that is substantially parallel to the longitudinal direction of the vehicle. The transverse coordinate axis X and the longitudinal coordinate axis Y intersect at the origin O.

[0043] The longitudinal direction of the vehicle 1 is, for example, a straight line passing through the center of the axle of the vehicle 1, the front of the vehicle, and the rear of the vehicle.

[0044] Preferably, the reference system 3 is an orthogonal coordinate system.

[0045] The lateral direction of the vehicle 1 is perpendicular to the longitudinal direction of the vehicle 1 and is, for example, a straight line passing through the left front door and passing through the right front door.

[0046] Advantageously, the origin O is located flush with the camera 2 of the vehicle 1. The camera 2 is a forward-looking camera located at the front of the vehicle 1.

[0047] The marking line 11 includes an initial point Pi, the coordinates of which in the coordinate system 3 are represented by an abscissa and an ordinate and can be determined based on the modeling data transmitted by the image processing device of the camera 2.

[0048] Advantageously, the abscissa and ordinate of the origin O are equal to zero. ​

[0049] Vehicle 1 includes a device 5 for detecting false alarms, which can determine the ordinate of the points on the marking lines 11 and 12 according to its abscissa.

[0050] In a preferred embodiment, the abscissa Xi of the initial point Pi is substantially equal to zero. Therefore, the abscissa Xi substantially coincides with the origin O. The ordinate of the initial point Pi is the first distance Yi from vehicle 1 to the marking line 11 at the first moment T1. The first distance Yi is the ordinate of the initial point Pi at the origin O. The first distance Yi can be determined according to the modeling data of the marking line 11. The device 5 for detecting false alarms can determine the first distance Yi according to the modeling data of the marking line 11 transmitted by the image processing device of the camera 2.

[0051] The marking line 11 further includes a horizon point Ph, and the abscissa of this horizon point at the first moment T1 is the first abscissa Xh corresponding to the distance that vehicle 1 will travel within a predefined horizon time dTh.

[0052] Vehicle 1 includes a speed sensor 6. The device 5 for detecting false alarms can collect the speed of the vehicle at a set moment from the speed sensor 6. According to this vehicle speed, the device for detecting false alarms can calculate the distance that vehicle 1 will travel within a predefined horizon time dTh.

[0053] The predefined horizon time dTh preferably ranges between 0.9 seconds and 1.1 seconds and is usually equal to 1 second.

[0054] For example, when the horizon time dTh is equal to one second and the vehicle speed is equal to 25 meters per second, the first abscissa Xh is equal to 25 meters.

[0055] The device 5 for detecting false alarms can determine the ordinate (referred to as the first ordinate Yh) of the horizon point Ph at the first moment T1 according to the pre-calculated first abscissa Xh and according to the modeling data of the marking line 11 transmitted by the image processing device of the camera 2.

[0056] Figure 2 Different from Figure 1 is that it shows vehicle 1 traveling in lane 10 at the second moment T2. The first moment T1 precedes the second moment T2. When vehicle 1 travels in lane 10, the vehicle travels a non-zero distance in lane 10 between the first moment T1 and the second moment T2.

[0057] The absolute value of the time difference DT between the first moment T1 and the second moment T2 preferably ranges between 1 millisecond and 150 milliseconds.

[0058] Since the reference system 3 is bound to the vehicle 1 and the position of the vehicle changes in the lane 10, at the second moment T2, the marking line 11 is at a second distance Y’i from the vehicle 1. The second distance Y’i of the vehicle 1 is the ordinate of the initial point P’i, so the abscissa Xi of this initial point coincides with the origin O, as Figure 1 shown in.

[0059] The marking line 11 also includes a new horizon point P’h, the abscissa of which at the second moment T2 is the second abscissa X’h of the second moment T2 corresponding to the distance that the vehicle 1 will travel within the predefined horizon time dTh.

[0060] The first abscissa Xh and the second abscissa X’h are not necessarily equal, but are very close. Specifically, the horizon time dTh is a fixed value, but the speeds of the vehicle 1 at the first moment T1 and the second moment T2 are not necessarily the same. However, the absolute value of the time difference DT between the first moment T1 and the second moment T2 is less than 150 milliseconds, so the difference between the first abscissa Xh and the second abscissa X’h is small.

[0061] For example, in the case where the horizon time dTh is equal to one second, if the difference in the speed of the vehicle between the first moment T1 and the second moment T2 is equal to 0.05 meters per second, the difference between the first abscissa Xh and the second abscissa X’h is equal to 0.05 meters.

[0062] Based on the speed of the vehicle at the second moment T2, the device 5 for detecting false alarms can calculate the distance that the vehicle 1 will travel within the predefined horizon time dTh, which corresponds to the second abscissa X’h.

[0063] The device 5 for detecting false alarms can determine the ordinate (referred to as the second ordinate Y’h) of the horizon point P’h at the second moment T2 based on the pre-calculated second abscissa X’h and the modeling data of the marking line 11 transmitted by the image processing device of the camera 2.

[0064] Figure 3 and Figure 4 respectively show Figure 1 and Figure 2 alternative embodiments of.

[0065] Figure 3 Differing from Figure 1 is that the abscissa Xi of the initial point Pi does not coincide with the origin O. The abscissa Xi is at a predefined distance from the origin O.

[0066] As Figure 1 shown in, at the first moment T1, the first distance Yi from the vehicle 1 to the marking line 11 is defined by the ordinate of the point Pi.

[0067] The device 5 for detecting false alarms can determine a first distance Yi based on the abscissa Xi (which is preset and known) and based on the modeling data of the marking line 11 transmitted by the image processing device of the camera 2.

[0068] Figure 4 Differing from Figure 2 is that the abscissa Xi of the initial point P’i does not coincide with the origin O. The abscissa Xi is at a predefined distance from the origin O that is the same as Figure 3 in

[0069] As in Figure 2 , at the second moment T2, the second distance Y’i from the vehicle 1 to the marking line 11 is defined by the ordinate of the point P’i.

[0070] The device 5 for detecting false alarms can determine a second distance Y’i based on the abscissa Xi (which is preset and known) and based on the modeling data of the marking line 11 transmitted by the image processing device of the camera 2.

[0071] As in Figure 3 and Figure 4 shown, in the case where the camera 2 is not a forward-looking camera located at the front of the vehicle, it may be advantageous to make the origin O not coincide with the abscissa Xi.

[0072] For example, the camera 2 is located on the roof of the vehicle 2. Advantageously, the abscissa Xi is the distance between the origin O located on the camera 2 and the front of the vehicle 2.

[0073] Figure 5 is Figure 2 a variant of Figure 4 which can also be applied to

[0074] Figure 5 Differing from Figure 2 is that the second ordinate Y’h at the second moment T2 is determined by the device 5 for detecting false alarms based on the calculated first abscissa Xh at the first moment T1. The advantage is that the first ordinate Yh and the second ordinate Y’h are determined using the same first abscissa Xh.

[0075] Figure 6 is an example of the image 20 taken by the camera 6. The coordinate system 3 is superimposed on the image 20 of the camera 6. The marking lines 11, 12 are discontinuous.

[0076] shows the origin Pi and the horizon point Ph located on the marking line 11 (such as Figure 1 shown).

[0077] Figure 7 shows a method for detecting false alarms of the image processing device of the camera 2 located on a vehicle 1 such as Figures 1 to 5 shown.

[0078] The method for detecting false alarms allows the detection of the following false alarms: initial discontinuity false alarm FPi, horizon discontinuity false alarm FPh, and range false alarm FPp.

[0079] To detect the initial discontinuity false alarm FPi, the method for detecting false alarms includes the following steps:

[0080] - A first step 101 of determining a first distance Yi of the vehicle 1 to the marking line 11 at a first moment T1 according to the modeling data of the marking line 11,

[0081] - A second step 102 of determining a second distance Y’i of the vehicle 1 to the marking line 11 at a second moment T2 according to the modeling data of the marking line 11,

[0082] - An initial discontinuity calculation step 103 of calculating an initial discontinuity Di, where the initial discontinuity Di is equal to the absolute value of the difference between the first distance Yi and the second distance Y’i,

[0083] - A step 104 of detecting an initial discontinuity false alarm, in which the initial discontinuity Di is compared with a preset low initial threshold TBi. If the initial discontinuity Di is greater than the low initial threshold TBi, an initial discontinuity false alarm FPi is detected.

[0084] In a preferred embodiment, the initial discontinuity Di is also compared with a preset high initial threshold THi. If the initial discontinuity Di is greater than the low initial threshold TBi and less than the high initial threshold THi, an initial discontinuity false alarm is detected. This allows false alarms not to be detected when the vehicle 1 changes lanes.

[0085] The low initial threshold TBi is equal to a value included between 0.1 meter and 0.3 meter, usually equal to 0.15 meter.

[0086] The high initial threshold THi is equal to a value included between 1.7 meters and 2.3 meters.

[0087] The absolute value of the time difference DT between the first moment T1 and the second moment T2 is a value less than or equal to the quotient of the low initial threshold TBi divided by a lateral speed substantially equal to two meters per second.

[0088] If the lateral speed is lower than two meters per second, the lateral movement is considered involuntary. If it exceeds this speed, the lateral movement is considered intentional. Therefore, choosing a speed of two meters per second is advantageous because it covers scenarios of involuntary lateral movement.

[0089] The absolute value of the time difference DT between the first moment T1 and the second moment T2 is greater than one millisecond, so as to give the device 5 for detecting false alarms sufficient time to execute the steps of the method for detecting false alarms.

[0090] Therefore, in the case where the low initial threshold TBi is equal to 0.1 meter, the absolute value of the time difference DT is a value included between 1 millisecond and 50 milliseconds, which is the result of dividing 0.1 by 2.

[0091] Therefore, in the case where the low initial threshold TBi is equal to 0.3 meter, the absolute value of the time difference DT is a value included between 1 millisecond and 150 milliseconds, which is the result of dividing 0.3 by 2.

[0092] To detect the false alarm FPh of the horizon discontinuity, the method for detecting false alarms includes the following steps:

[0093] - A third step 201 of determining the first ordinate Yh of the horizon point Ph of the marking line 11 at the first moment T1 according to the modeling data of the marking line 11,

[0094] - A fourth step 202 of determining the second ordinate Y’h of the horizon point P’h of the marking line 11 at the second moment T2 according to the modeling data of the marking line 11,

[0095] - A horizon discontinuity calculation step 203 of calculating the horizon discontinuity Dh, where the horizon discontinuity Dh is equal to the absolute value of the difference between the first ordinate Yh and the second ordinate Yh’ minus the initial discontinuity Di,

[0096] - A step 204 of detecting the false alarm of the horizon discontinuity, in which the horizon discontinuity Dh is compared with a preset horizon threshold THh, and if the horizon discontinuity Dh is greater than the horizon threshold THh, the false alarm FPh of the horizon discontinuity is detected.

[0097] The first ordinate Yh of the horizon point Ph and the second ordinate Y’h of the horizon point P’h are determined according to the first abscissa Xh or the second abscissa X’h calculated according to the modeling data of the marking line 11 as explained in the description of Figure 2 、 Figure 4 and Figure 5 .

[0098] The horizon threshold THh is included between 0.2 meter and 1 meter.

[0099] The modeling data includes the optimal viewing distance Xp.

[0100] Figure 6An example of optimal viewing distance Xp is given in FIG. Truck 21 is traveling in lane 10. For camera 6, truck 21 is an obstacle in lane 10. Truck 21 partially obscures marking lines 11 and 12. In this example, optimal viewing distance Xp substantially corresponds to the distance between truck 21 and camera 6.

[0101] In order to detect a range false alarm FPp, the method for detecting a false alarm further includes a step 304 of detecting a range false alarm, in which the best viewing distance Xp is compared with a range threshold THp, the range threshold being equal to the distance that the vehicle will travel within a predefined range time dTp, and if the best viewing distance Xp is less than the range threshold THp, a range false alarm FPp is detected.

[0102] Advantageously, the step 304 of detecting a range false alarm is performed at the second time T1 and / or the second time T2.

[0103] The predefined range time dTp is a value comprised between 0.8 seconds and 0.99 seconds, typically equal to 0.95 seconds.

[0104] For example, where the range time dTp is equal to 0.95 seconds and the vehicle speed is equal to 25 meters per second, the range threshold THp is equal to 23.75 meters, which is 0.95 multiplied by 25.

[0105] Of course, this is considered equivalent to converting the optimal visibility distance Xp into the time required for the vehicle 1 to travel the optimal visibility distance Xp and comparing the time with the range threshold THp.

[0106] If any one of the initial discontinuity false alarm FPi, the horizon discontinuity false alarm FPh, and the range false alarm FPp is detected, a false alarm is detected.

[0107] Advantageously, if a false alarm is detected, the method for detecting a false alarm further comprises a step 124 of displaying a visual warning to the user of the vehicle 1 in order to warn the user that the camera 2 temporarily cannot detect the marking line 11 .

[0108] For example, the vehicle 1 includes a display screen 100 on which marking lines 11, 12 of the vehicle 10 are represented. If the device 5 for detecting a false alarm detects a false alarm of the marking line 11, the representation 110 of the marking line 11 on the display screen 100 is modified, for example by changing the color. If no false alarm of the marking line 12 is detected, the representation 120 of the marking line 12 is not modified.

[0109] exist Figure 8a In FIG, no false alarms are detected and the representations 110 , 120 of the two marker lines 11 , 12 are black.

[0110] exist Figure 8bAmong them, a false alarm of the marking line 11 has been detected. The representation 110 of the marking line 11 is gray, and the representation 120 of the marking line 12 remains unchanged.

[0111] The visual warning embodiments are non-limiting. The visual warning can be a pictogram, an indicator light, etc.

[0112] The device 5 for detecting false alarms includes a program, which includes program instructions for implementing the steps of a method for detecting false alarms.

[0113] The device 5 for detecting false alarms is preferably connected to the speed sensor 6 of the vehicle 1 and the image processing device of the camera 2 through a wired link. The connection between the speed sensor 6 and the device 5 for detecting false alarms can be direct or indirect. For example, an intermediate processor can be located between the speed sensor 6 and the device 5 for detecting false alarms, and this intermediate processor enables data to be filtered and verified before being transmitted to the device 5 for detecting false alarms.

[0114] The communication protocol adopted between the speed sensor 6 and the device 5 for detecting false alarms is, for example, the CAN protocol.

[0115] The communication protocol adopted between the camera 2 and the device 5 for detecting false alarms is, for example, the CAN protocol.

[0116] Advantageously, the program instructions for implementing the steps of the method for detecting false alarms are cyclically executed at a period equal to the time difference DT. When the program is executed for the first time, the first moment T1 corresponds to a certain moment in the first time period, and the second moment T2 corresponds to a certain moment in the second time period. When the program is executed for the second time, the first moment T1 corresponds to the moment in the second time period, and the second moment T2 corresponds to a certain moment in the third time period, and so on.

[0117] The above examples and descriptions are given taking the marking line 11 as an example. By analogy, the same examples and explanations also apply to the marking line 12.

Claims

1. A method for detecting false alarms of an image processing device of a camera (2) located on a vehicle (1), the image processing device being capable of transmitting modeling data of a marking line of a lane (10) of the vehicle (1) in a reference system (3) bound to the vehicle (1), the method for detecting false alarms comprising the following steps: - A first step (101) of determining a first distance (Yi) from the vehicle (1) to the marking line at a first moment (T1) based on the modeling data of the marking line, - A second step (102) of determining a second distance (Y’i) from the vehicle (1) to the marking line at a second moment (T2) based on the modeling data of the marking line, - An initial discontinuity calculation step (103) of calculating an initial discontinuity (Di), the initial discontinuity (Di) being equal to the absolute value of the difference between the first distance (Yi) and the second distance (Y’i), - A step (104) of detecting an initial discontinuity false alarm, in which the initial discontinuity (Di) is compared with a preset low initial threshold (TBi), if the initial discontinuity (Di) is greater than the low initial threshold (TBi), an initial discontinuity false alarm (FPi) is detected, the low initial threshold (TBi) being equal to a value included between 0.1 m and 0.3 m, the time difference (DT) between the first moment (T1) and the second moment (T2) being less than or equal to the quotient of the low initial threshold (TBi) divided by a lateral speed equal to two meters per second, and the time difference (DT) being greater than one millisecond, - The initial discontinuity (Di) is also compared with a preset high initial threshold (THi), if the initial discontinuity (Di) is greater than the low initial threshold (TBi) and less than the high initial threshold (THi), an initial discontinuity false alarm is detected.

2. The method for detecting false alarms according to claim 1, the high initial threshold (THi) being equal to a value included between 1.7 m and 2.3 m.

3. The method for detecting false alarms according to claim 1 or 2, the reference system (3) comprising a longitudinal coordinate axis (Y) parallel to the lateral direction of the vehicle (1) and an origin (O) flush with the camera (2) of the vehicle (1), the longitudinal coordinate axis (Y) intersecting the marking line at intersection points on the same side at the first moment (T1) and the second moment (T2) respectively, the first distance (Yi) and the second distance (Y’i) corresponding to the ordinates of the intersection points relative to the origin (O).

4. The method for detecting false alarms according to claim 3, the reference system (3) comprising a transverse coordinate axis (X) parallel to the longitudinal direction of the vehicle (1), the method for detecting false alarms further comprising the following steps: - A third step (201) of determining a first ordinate (Yh) of a horizon point (Ph) of the marking line based on the modeling data of the marking line, the first abscissa (Xh) of the horizon point at the first moment (T1) being a distance corresponding to the distance that the vehicle (1) will travel within a predefined horizon time (dTh), - A fourth step (202) of determining a second ordinate (Y’h) of a horizon point (P’h) of the marker line based on the modeling data of the marker line, wherein a second abscissa (X’h) of the horizon point at the second moment (T2) corresponds to a distance that the vehicle (1) will travel within the predefined horizon time (dTh). - A horizon discontinuity calculation step (203) of calculating a horizon discontinuity (Dh), where the horizon discontinuity (Dh) is equal to the absolute value of the difference between the first ordinate (Yh) and the second ordinate (Yh’) minus the initial discontinuity (Di). - A step (204) of detecting a false alarm of horizon discontinuity, in which the horizon discontinuity (Dh) is compared with a preset horizon threshold (THh). If the horizon discontinuity (Dh) is greater than the horizon threshold (THh), a false alarm of horizon discontinuity (FPh) is detected.

5. The method for detecting false alarms according to claim 4, wherein the predefined horizon time (dTh) is a value included between 0.9 seconds and 1.1 seconds, and the horizon threshold (THh) is a value included between 0.2 meters and 1 meter.

6. The method for detecting false alarms according to claim 1 or 2, wherein the modeling data includes an optimal visible distance (Xp), and the method for detecting false alarms further includes a step (304) of detecting a range false alarm, in which the optimal visible distance (Xp) is compared with a range threshold (THp) equal to the distance that the vehicle will travel within a predefined range time (dTp). If the optimal visible distance (Xp) is less than the range threshold (THp), a range false alarm (FPp) is detected.

7. The method for detecting false alarms according to claim 6, wherein the predefined range time (dTp) is a value included between 0.8 seconds and 0.99 seconds.

8. The method for detecting false alarms according to claim 1 or 2 further includes the following step (124): If a false alarm among the initial discontinuity false alarm (FPi), the horizon discontinuity false alarm (FPh), and the range false alarm (FPp) is detected, a visual warning is displayed to the user of the vehicle (1) to warn the user that the camera (2) cannot detect the marker line temporarily.

9. A computer program product including program instructions, which, when executed by a computer, implement the steps of the method for detecting false alarms according to any one of claims 1 - 8.

10. A readable data medium storing the computer program product according to claim 9.

Images