Method for detecting flashing lights

Abstract

A method for detecting flashing lights, implemented by a device comprising at least one image acquisition device and a computer, characterized in that it comprises: an acquisition (100) of an image stream; for at least one image of said image stream, a segmentation (120) of the image comprising: a calculation (121) of a differential image resulting from the pixel-by-pixel difference in intensity between the image under consideration and a preceding image in the stream, as well as an extraction (122) of the segments of the differential image exhibiting absolute values ​​of intensity difference greater than a predefined threshold; and for each of the segments of said segmentation, an association (130) of the segment with a flashing light tracking marker in the image stream. Abstract figure: Figure 1

Description

Title of the invention: Method for detecting flashing lights technical field

[0001] This disclosure falls within the field of image processing. The invention relates more particularly to a method for detecting flashing lights. Prior art

[0002] When a driver is on the road, they may encounter unusual situations. These situations, which fall outside the driver's everyday routine and habits, often lead to dangerous and unpredictable behavior. To draw drivers' attention in particular, the highway code includes the use of flashing warning lights.

[0003] Such warning devices are found, for example, on fixed elements of the environment such as traffic lights (when the yellow light flashes to warn of a dangerous situation, such as an intersection), alongside certain traffic signs (particularly for dangerous pedestrian crossings), or in work zones to draw attention to temporary signage. They are also found on moving elements such as any vehicle with turn signals, emergency vehicles with the flashing beacon of police cars, ambulances, or fire trucks, or in routine use, such as the flashing beacon of agricultural machinery.

[0004] In the prior art, it is known to equip vehicles with devices capable of recognizing elements of the roadway, such as signage or vehicles, in particular to provide road users with tools enabling them to improve safety, and in particular by equipping autonomous or semi-autonomous vehicles in order to improve their efficiency.

[0005] It is thus known technically that methods for detecting potential light sources are based on segmenting an image according to a pixel intensity threshold, and, depending on the application, on the intensity of certain pixel color components relative to other components. These methods, however, have certain limitations: they are only sufficiently effective with lights of constant intensity and / or lights in a fixed position, such as traffic lights.

[0006] When the lights are flashing and / or follow different patterns of intensity variation, these methods are inefficient and generate a large number of errors. Furthermore, when the lights are moving, it is very difficult to maintain tracking of flashing lights when they are in a switched-off phase. In addition, the saturation of a distant flashing light decreases, rendering the methods inoperative. methods based on the intensity of a color component of pixels and those based on the light intensity of pixels.

[0007] The methods of the prior art are therefore insufficient to effectively detect flashing lights, for example those of emergency vehicles, which are generally used to prevent dangerous situations. There is a need for a solution that covers these situations. Summary

[0008] This disclosure improves the situation.

[0009] A method for detecting flashing lights is proposed, implemented by a device comprising at least one image acquisition device and a computer, further comprising: - the acquisition of an image stream comprising a plurality of successively acquired images forming said image stream, - for at least one image of said image stream, • an image segmentation including: • a calculation of a differential image resulting from the difference in intensity, pixel by pixel, between the image under consideration and an image that precedes it in the stream, • Extraction of segments from the differential image exhibiting absolute intensity difference values ​​exceeding a predefined threshold, • for each segment of said segmentation, an association of the segment with a blinking light tracking marker in the image stream, said tracking marker being determined as follows: • if there is a tracking marker at a distance from the segment in question that is less than a predefined threshold, said segment is associated with said tracking marker, and the position of said marker is updated to that of the segment, and • otherwise, a tracking marker is defined in the image stream at the position of the segment under consideration and is associated with it.

[0010] In a variant of the flashing light detection method, the association of a segment with a tracking marker further includes a check that the sequence of segments associated with the tracking marker, including the segment in question, respects a predefined pattern, for example an alternation of rising and falling edges of segment intensity.

[0011] In another variant, the method for detecting flashing lights further includes a classification of tracking markers.

[0012] Alternatively, the tracking markers are classified among different types of lights.

[0013] Optionally, the different types of lights include flashing lights with different patterns, frequencies, and intensities, as well as steady lights.

[0014] In a variant of the flashing light detection method, the classification of each tracking marker is performed by analyzing a pattern formed by the sequences of segments associated with the marker

[0015] Alternatively, the classification of each tracking marker is performed by a machine learning classification algorithm, pre-trained on a training database comprising a plurality of image streams of different types of lights such as flashing lights following different patterns, different frequencies, and different intensities

[0016] In one embodiment, the method for detecting flashing lights further comprises, for each acquisition of an image of said image stream: - a calculation of the displacement, of a given image relative to the preceding image in the stream, of elements present in said images, - updating the position of at least one tracking marker, based on the calculated displacement.

[0017] In one variant, the flashing light detection method further includes the removal of tracking markers that have not been associated with any segments extracted from a number of images greater than a predefined limit.

[0018] In one embodiment, the method for detecting flashing lights further comprises, for each acquisition of an image of the flow: - a definition, prior to said segmentation, of an area of ​​interest on the image under consideration, and - the implementation of image segmentation only on the area of ​​interest,

[0019] In one embodiment, the image acquisition device is mounted in a vehicle, and the area of ​​interest is determined from the points at infinity and vanishing lines of the image, as the area comprising the substantially fixed elements from one image to another of the image stream.

[0020] In one variant of the flashing light detection method, the definition of the area of ​​interest further includes: - the evaluation of the distance of certain elements present in the image, to the image acquisition device, and - the exclusion of areas of the image containing elements at a distance thus evaluated less than a predefined limit distance.

[0021] Alternatively, the flashing light detection method further includes detection and tracking of dynamic objects and the area of ​​interest includes the pixels associated with detected and tracked dynamic objects.

[0022] According to another aspect, a computer program product is proposed, comprising code instructions for implementing the method of detecting flashing lights, when executed by a computer.

[0023] According to another aspect, a flashing light detection device is proposed comprising an image acquisition device and an embedded or remote computer configured to implement the flashing light detection method.

[0024] According to a final aspect, a motor vehicle is proposed comprising a flashing light detection device. Brief description of the drawings

[0025] Other features, details and advantages will become apparent from reading the detailed description below and from analyzing the accompanying drawings, in which: Fig. 1

[0026] [Fig.1] Fig.1 shows a schematic representation of the process according to one embodiment. Fig. 2

[0027] [Fig.2] Fig.2 shows a schematic representation of a vehicle implementing the process according to one embodiment. Fig. 3

[0028] [Fig. 3] [Fig. 3] Schematically illustrates successive stages of the process, according to one embodiment thereof. Description of embodiments

[0029] For example, to provide a safety tool or to make it autonomous for detecting dangerous situations, a vehicle must be able to recognize flashing lights it encounters on the road. Reference is now made to [Fig. 1], which schematically represents a flowchart of the process for detecting flashing lights according to one embodiment, and to [Fig. 2], which represents a vehicle according to one embodiment of the process.

[0030] This method is implemented by a flashing light detection device 200, comprising at least one computer 201, for example of the processor, microprocessor, microcontroller, field-programmable gate array (FPGA, etc.), and / or specialized integrated circuit (ASIC, etc.) type. The computer 201 is advantageously located onboard in the vehicle 210. Alternatively, the computer 201 may be located remotely from the vehicle and be in remote communication with an on-board communication device 203 and / or a vehicle component 204.

[0031] For example, the communication device 203 is a device following the V2X (Vehicle-to-everything) vehicular communication technology and particularly the V2N (Vehicle-to-Network) technology.

[0032] The device may further include a memory 202, storing code instructions executed by the computer 201 to implement the method. The memory 202 may also store one or more pre-trained models, described in more detail below, to determine in greater detail the types of lights detected. The memory 202 may be a non-volatile electronic memory, an optical hard drive, etc.

[0033] The vehicle 210 further includes an image acquisition device 204, enabling the computer 201 to receive images for the implementation of the method. This device 204 may be an on-board camera or an on-board still camera.

[0034] In order to study the flashing lights in the vehicle's environment, the method first comprises, in one embodiment, an acquisition 100 of an image stream. This acquisition 100 can be carried out using the image acquisition device 204 mounted on the vehicle, for example from a video stream or a series of successive photos.

[0035] In one embodiment, this acquisition 100 includes image preprocessing, depending on the application. Certain pixel components can be removed or modified so that the images are advantageously suited to the application. For example, to detect only red flashing lights, the green and blue components of the image pixels are removed. In another example, the images are advantageously converted to shades of gray to reduce memory and computational load.

[0036] Acquiring such a stream of images allows one to advantageously observe the evolution over time of certain elements present in the image. In particular, the blinking of a light is a phenomenon that unfolds over time.

[0037] Thus, the method comprises comparing an image in the stream with a preceding image in the stream. The preceding image is not necessarily the one immediately preceding it in the stream; several other images may separate them. Furthermore, the image stream may be pre-sampled so as, advantageously, not to contain images that separate the images to be compared.

[0038] This comparison is carried out by calculating 121 a so-called differential image, because it results from the difference, pixel by pixel, between the two images. Such a differential image expresses the variations that each pixel has undergone during the time that separates the capture of each of the images. Consequently, it advantageously contains information on the rising and falling fronts of flashing lights.

[0039] The differential image Imdiff can thus be constructed in the following manner:

[0040] Im^ ff- Jmn - Imnk, with Imn the nth image considered in the stream and k the number of images which separates the image being considered from the one with which it is compared and which precedes it in the flow.

[0041] Generally, the acquisition 100 of the images of the stream by such a device 204 follows a constant acquisition period, and we can then also note:

[0042] Imdiff = Imt-lmtkT, with t the time at which the image in question is acquired and T the acquisition period.

[0043] The difference between two images is understood to be the difference, for each pixel of the first image, between the components of the pixel in question and the respective components of the corresponding pixel (at the same position) in the second image. The result of such a difference can advantageously lead to positive or negative component values ​​for the pixels of the differential image. For example, if the red component of a pixel is negative, this indicates a decrease in the red level for that pixel during the time between the acquisition of the two images.

[0044] The components or color components are the values ​​that allow the color of a pixel to be expressed, for example the amount of red, green and blue, or the hue, saturation and vibrance. In variations, this can also correspond to the gray level of the pixel, its luminance, etc.

[0045] The differential image can thus express the intensity variations of the pixels. The intensity variations considered may depend on the application case and may include: variations in the overall intensity of the pixel, those of one or more of its color components such as a color variation from red to blue, or a combination of variations in the overall intensity and color of the pixel.

[0046] The number of images k that separates the image Imn from the image Imnk can be constant, for example equal to 1, so as to compare two images acquired consecutively.

[0047] In one embodiment, the number k can also follow a variable function, for example a triangular or sawtooth function. Indeed, for a constant k, a flashing light with a period that is a multiple of kT is in phase with the calculation 121 of the differential image: the difference is then always made between two images in which the flashing light is in the same state (either always on or always off), and essentially no variation is then measured on the differential image. A variable number of separation images k advantageously eliminates this risk.

[0048] In one embodiment, the calculation of the differential image includes a realignment of the image Imn with respect to the image Imnk taking into account the displacement from one image to the other of elements present on the two images.

[0049] The method then comprises extracting 122 areas of the differential image, called segments, when the absolute intensity values ​​of the pixels in the area are greater than a predefined threshold. For example, the threshold is a variation of + / - 20%. The predefined value for the threshold can advantageously be adapted to each application case.

[0050] Depending on the application, the intensity of a segment is understood to be an average (or a median, etc.) of the intensity of the pixels that make up that segment. This intensity may relate to a particular component of the pixels of the segment, or to a combination of components, or to the result of a calculation on a component or a combination of components, such as, for example, the luminance, saturation, or gray level of the pixels of the segment.

[0051] In one variant, the extracted area is approximated by a geometric shape, for example elliptical or preferably rectangular, so as to be able to advantageously easily define the position of a segment by a remarkable point (center or corner) rather than by a barycenter of the pixels of any shape.

[0052] The segmentation 120 of the process, which includes the calculation 121 of the differential image and the extraction 122 of the segments, makes it possible to establish a set of segments which each correspond to a strong variation (defined by the threshold), and which are thus potential candidates to be evidence of the presence of one or more flashing lights in the image.

[0053] In one embodiment, the process advantageously includes, prior to the segmentation 120, the definition 110 of an area of ​​interest on the image, to which the segmentation 120 is limited.

[0054] Indeed, the extracted segments correspond to areas whose pixels have undergone significant variations. However, when the vehicle is moving with the onboard image acquisition device 204, the closer an element of the image Imn k was physically to the vehicle, the more likely it is to have a different position in the image Imn. This difference in position can lead to major variations in the pixels and therefore the extraction of false-positive segments associated with these variations.

[0055] Defining an area of ​​interest in the image, excluding elements too close to the vehicle, to limit segmentation advantageously avoids the extraction of segments corresponding to the movement of nearby elements. The segmentation thus advantageously has better recall (or better sensitivity).

[0056] The definition of the area of ​​interest is generally achieved by applying a filter to the images, so as to apply a multiplicative weight (or coefficient) of 0 to the (components of the) pixels outside the area of ​​interest and of 1 to the pixels inside the area of ​​interest.

[0057] In an unillustrated embodiment, said filter includes weights between 0 and 1, so the delimitation of the area of ​​interest is not clear. For example, the area of ​​interest may extend around an interest point, so that the further a pixel is from the interest point, the lower its weight, following a linear, polynomial, exponential or logarithmic decrease, etc.

[0058] In one embodiment, the area of ​​interest can be defined as including a set of points considered at infinity with respect to the image acquisition device.

[0059] The delimitation of the area of ​​interest on the image can be done using the vanishing lines (and the vanishing point or point at infinity that can be deduced from them), which advantageously allow the determination of the relative distance between elements of the image with respect to the image acquisition device that originated the image. In particular, a set of points located at a distance less than a predefined distance threshold from the image's vanishing point can be approximated as points located at infinity.

[0060] Generally, the vanishing point is located in the central part of the image, towards which the vanishing lines are directed. The delimitation of the area of ​​interest can thus be fixed beforehand (identical, regardless of the image), as can all the pixels in the central part. For example, it can be in the form of an oval encompassing all the pixels in the central part. Such a delimitation advantageously reduces the computation time required to define the area of ​​interest.

[0061] However, the area of ​​interest can also vary more or less depending on the image to which it is applied. In one variant, the area of ​​interest is delimited based on a predefined distance threshold of the image elements from the image acquisition device: elements located at a distance less than the predefined threshold define the area of ​​interest. Such delimitation makes it possible to obtain an area of ​​interest that is very faithful to reality. In addition, the computer 201 can advantageously perform 3D geometry recognition of the object in order to refine the evaluation of its distance from the image acquisition device 204.

[0062] In one embodiment, the computer 201 detects and tracks dynamic objects in the image stream. For example, the computer implements a Kalman filter on the images. The area of ​​interest can advantageously include the pixels of the dynamic objects, so that, for example, vehicles likely to emit flashing lights as emergency vehicles are part of the area of ​​interest.

[0063] In another embodiment, the area of ​​interest is of a predefined shape, for example, preferably an ellipse, whose geometric characteristics (in the example, the semi-major axis, the semi-minor axis, and the center) can vary according to the image or pair of images. The center of the ellipse can advantageously be the vanishing point of the image.

[0064] The dimensions of the area of ​​interest can also take into account the movement of the vehicle equipping the image acquisition device between images or the speed of the vehicle. For example, the slower the vehicle 210 moves between the two images, the larger the area of ​​interest can be.

[0065] Thus, this last variant advantageously allows obtaining an effective area of ​​interest, without requiring too much computing power and time.

[0066] Alternatively, in this latter variant or in a variant where the area of ​​interest is predetermined, the delimitation includes a subsequent check. This check excludes foreground elements of the image that are both close to the image acquisition device 210 and close to the vanishing point in the image. For example, this element is a vehicle following the vehicle 210 equipped with the image acquisition device 200. This check advantageously makes it possible to exclude elements that might initially be included in the area of ​​interest due to their proximity to the vanishing point in the image, despite their proximity to the device 210. These elements can be detected using known methods for detecting road features, and in particular for detecting vehicles.

[0067] The method then includes associating 130 of the segments with tracking markers for a flashing light. A tracking marker makes it possible to maintain the tracking of a flashing light from one image to the next of the stream. Two cases are possible for a segment thus extracted during segmentation: either a pre-existing tracking marker is close (i.e., less than a predefined distance threshold depending on the application) to this segment, or the segment is isolated (no tracking marker is close to it).

[0068] An isolated segment corresponds to a sudden change (i.e., a rising or falling edge) in pixel intensity in a part of the image that had not previously changed: it is a candidate for the appearance of a new blinking light in the image. Consequently, association 130 includes the creation of a new tracking marker at the position of the isolated segment and associates itself with it.

[0069] A segment close to a marker is a candidate for the continuation of the blinking of a light already segmented in previous images. The method therefore advantageously includes associating this segment with this marker. If a segment is close to several markers, it is associated with the nearest marker. When a segment is associated with a pre-existing marker, the association includes updating the position of the tracking marker to the position of the segment associated with it, so that a marker successively associated with a sequence of segments from a moving light can synchronize its position with that of the moving light.

[0070] In one embodiment, the association includes a check that the associated segment falls within a predefined pattern corresponding to a blink. For example, it is checked that a marker previously associated with a segment linked to a light that turns on (positive sign of intensity variation in the pixel components of the segment on the differential image) is associated with a segment linked to a light that turns off (negative sign of intensity variation). The pattern is then an alternation of rising and falling intensity edges in the segments of the sequence.

[0071] Alternatively, the predefined pattern can also relate to the intensity of the associated segments, so that the evolution of the intensity follows a blinking pattern.

[0072] In one embodiment of the method, the position of markers associated with no segment is updated 140. A moving blinking light in the image stream continues to move when it does not generate variations / segments (stable intensity). Thus, the update 140 includes calculating the displacement of the elements of image Imn relative to image Imnk, and from this calculation, estimating the displacement of the blinking light linked to the tracking marker.

[0073] These elements may be characteristic features of the roadway present in both images, such as a lamppost that has moved from one image to the other. Alternatively, the elements may be a simple characteristic arrangement of pixels of particular intensities, when this arrangement is present in both images at different positions.

[0074] Taking into account a plurality of elements makes it advantageous to estimate a precise displacement.

[0075] The position of the tracking marker is thus updated by following this movement.

[0076] In one variant, the estimated displacement is zero because the images Imn and Imnk are identical and the vehicle is stationary. Advantageously, updating the tracking markers is avoided in this particular case.

[0077] Alternatively, this adjustment also takes into account a plurality of previous images, allowing the displacement to be estimated, advantageously in a more precise manner, by correlating it with the displacements of the previous images.

[0078] Alternatively, the method may include detecting image elements using known methods, and advantageously updating 140 tracking markers relative to the detected elements. For example, a tracking marker corresponding to a flashing light on a detected vehicle can be advantageously updated 140 depending on the movement made by the detected vehicle.

[0079] Once the marker's position has been updated by following the estimated displacement, the marker is theoretically located at the position of the light with stable intensity. Thus, if the light generates a new segment on a subsequent image, the correct tracking marker (corresponding to the same flashing light) will advantageously already be positioned close to the segment.

[0080] Advantageously, this update allows a marker to follow a flashing light, including when it is in a stable off phase.

[0081] Since a marker associated with no segment for a long period signifies the disappearance of a flashing light, updating a marker advantageously includes its deletion. Thus, according to one embodiment of the method, a marker associated with no segment for a predefined number of frames or time is deleted.

[0082] Advantageously, in one embodiment, this predefined number of images (or time) corresponds to a time greater than the blinking period of the predefined pattern of the process verification, so that a marker is not deleted by mistake for a blinking light whose blinking period is too high.

[0083] In another embodiment, the method includes removing a marker which, once its position is updated, is located outside the area of ​​interest for a predefined number of frames or time. Since no segment is created outside the area of ​​interest, the marker corresponds to a blinking light that moves out of the image field. The predefined time advantageously corresponds to a waiting period, during which the blinking light moving out of the image can change direction and re-enter the area of ​​interest.

[0084] In one embodiment, the method includes a classification 150 of tracking markers that determines the type of light to which the tracking marker is linked. This classification can advantageously differentiate the lights by their physical characteristics such as their blinking period, their blinking pattern, the light frequency or frequencies (colors) emitted, etc.

[0085] Such an analysis for classification 150 also advantageously allows the detection of a false positive, when a flashing light tracking marker has been associated with a non-flashing light.

[0086] In one embodiment, such a classification 150 is carried out by studying the sequence of segment associations with the marker under study. The method thus deduces from the sequence of segment associations with the tracking marker, the blinking pattern of the light associated with the marker and the remaining physical characteristics of the associated light.

[0087] In one embodiment, this classification 150 differentiates lights by their application. This classification 150 is advantageously more utilitarian because it allows one to determine which event the light corresponds to. For example, the classification determines whether the marker is following a traffic light, an ambulance beacon, a police beacon, etc.

[0088] Advantageously, in one embodiment, such a classification 150 is performed by a machine learning algorithm pre-trained on a set of labeled tracking markers (i.e., the type of light associated with the marker is known). For example, the training database includes tracking markers of flashing traffic lights, beacons of various emergency vehicles, etc.

[0089] Alternatively, the training database for classification 150 also includes flashing lights from different countries so that the algorithm can be advantageously effective regardless of the country of origin of the lights. Such an algorithm is, for example, capable of differentiating the light coming from a French ambulance beacon from an American ambulance beacon.

[0090] In one embodiment, the classification 150 is performed by the on-board computer 201 on the vehicle. Alternatively, it is performed by a remote computer; in this case, the vehicle includes a communication device 203 which transmits to a remote server 213 the information necessary for the classification (the set of tracking markers) and the remote server 213, after performing the classification calculation, transmits the results to the vehicle.

[0091] Reference is now made to [Fig.3] which illustrates the implementation of the different stages of the process according to one embodiment.

[0092] In this embodiment, the computer 201 receives the image 300 from the image acquisition device 204. This image includes an ambulance whose first beacon 301 has just turned on, and the second beacon 302 remains off (that is to say, it was already off during the previous segmentation and is still off).

[0093] Image 310 illustrates the definition 110 of an area of ​​interest 311 from the vanishing lines present on image 300. In this application case, a fixed elliptical shape and size are predefined, and the ellipse is centered on the vanishing point (at infinity) of the image.

[0094] Furthermore, this definition includes the exclusion of the area 312 comprising an element in the foreground of the image (a car) which is both close to the vanishing point of the image and physically close to the image acquisition device 204.

[0095] Image 320 is the differential image resulting from the difference 121 between image 300 and the preceding image (not shown) in the image stream of the image acquisition device 204. The black areas represent slight variations while that the white zone 321 (located at the level of the first 301 gyro beacon) represents an area of ​​strong variations in intensity (in absolute value).

[0096] Image 330 thus illustrates the result of the segmentation 120 of image 300. A new segment 333 is extracted 122 from area 321. This segment 333 is associated 130 with the first marker 332a, which is the marker 332 closest to segment 333.

[0097] Since the segment corresponds to a light that turns on, the method also includes verifying that the last segment associated with marker 332a was a segment corresponding to a light that turned off, in order to follow a blinking pattern with the new segment 333.

[0098] The second marker 332b, resulting from a previous segmentation (of the second beacon 302 of the ambulance), is not associated with any segment. Its position is nevertheless updated 140 to be located at the position where the light of the second beacon of the ambulance is estimated to reappear.

[0099] The process then includes a classification 150 of marker 332a, which, by its pattern of associations with segments from previous segmentations, is recognized as being linked to an ambulance flashing light.

Claims

1.

2. Demands A method for detecting flashing lights, implemented by a device (200) comprising at least one image acquisition device (204) and a computer (201), characterized in that it comprises: - an acquisition (100) of an image stream comprising a plurality of successively acquired images forming said image stream, - for at least one image of said image stream, • a segmentation (120) of the image comprising: • a calculation (121) of a differential image resulting from the difference in intensity, pixel by pixel, between the image under consideration and an image preceding it in the stream, • an extraction (122) of the segments of the differential image exhibiting absolute intensity difference values ​​greater than a predefined threshold, • for each of the segments of said segmentation, an association (130) of the segment to a blinking light tracking marker in the image stream, said tracking marker being determined as follows: • if there is a tracking marker at a distance from the segment in question less than a predefined threshold, said segment is associated with said tracking marker, and the position of said marker is updated (140) to that of the segment, and • otherwise, a tracking marker is defined in the image stream at the position of the segment under consideration and is associated with it. Method for detecting flashing lights according to claim 1 wherein the association (130) of a segment with a tracking marker further includes a verification that the sequence of segments associated with the tracking marker, including the segment in question, respects a predefined pattern.

3. Method for detecting flashing lights according to any one of the preceding claims further comprising a classification (150) of tracking markers.

4. A method for detecting flashing lights according to the preceding claim, wherein the tracking markers are classified among different types of lights.

5. A method for detecting flashing lights according to the preceding claim, wherein the different types of lights include flashing lights with different patterns, frequencies, and intensities, as well as steady lights.

6. Method for detecting flashing lights according to any one of claims 3 to 5, the classification (150) of each tracking marker is carried out by analyzing a pattern formed by the sequences of segments associated with the marker.

7. A method for detecting flashing lights according to any one of claims 3 to 6, wherein the classification (150) of each tracking marker is carried out by a machine learning classification algorithm, pre-trained on a training database comprising a plurality of image streams of different types of lights such as flashing lights following different patterns, different frequencies, and different intensities.

8. A method for detecting flashing lights according to any one of the preceding claims further comprising, for each acquisition of an image of said image stream: - a calculation of the displacement, of a considered image relative to the image preceding it in the stream, of elements present in said images, - the updating of the position of at least one tracking marker, from the calculated displacement.

9. A method for detecting flashing lights according to any one of the preceding claims, further comprising the removal of tracking markers that have not been associated with any segments extracted from a number of images exceeding a predefined limit.

10. A method for detecting flashing lights according to any one of the preceding claims, further comprising, for each acquisition of an image of the flow: - a definition (110), prior to said segmentation, of an area of ​​interest on the image considered, and - the implementation of the segmentation (120) of the image only on the area of ​​interest.

11. A method for detecting flashing lights according to the preceding claim, wherein the image acquisition device (204) is mounted in a vehicle (210), and the area of ​​interest is determined from the points at infinity and vanishing lines of the image, as the area comprising the substantially fixed elements from one image to the next of the image stream.

12. A method for detecting flashing lights according to any one of the preceding claims, wherein the definition of the area of ​​interest further includes: - evaluating the distance of the elements present in the image to the image acquisition device, and - excluding areas of the image comprising elements at a distance thus evaluated less than a predefined limit distance.

13. Method for detecting flashing lights according to claim 11 or 12 further comprising detection and tracking of dynamic objects and the area of ​​interest comprises the pixels associated with detected and tracked dynamic objects.

14. Product computer program, comprising code instructions for implementing the method according to any one of the preceding claims, when executed by a computer (201).

15. Flashing light detection device (200) comprising an image acquisition device (204) and an on-board or remote computer (201) configured to implement the flashing light detection method according to any one of claims 1 to 13.

16. Motor vehicle (210) comprising a flashing light detection device (200) according to the preceding claim.

Images