JP6596889B2 - Object detection device - Google Patents

Description

  The present invention relates to an object detection device that detects a series of reflectors that are present around a vehicle travel path and are composed of a plurality of reflectors arranged in a continuous manner.

  As an object detection device that detects an object existing around a vehicle, for example, an apparatus disclosed in Patent Document 1 is known. In Patent Document 1, it is described that the light reflected forward from the light projecting unit increases the beam intensity as it corresponds to the farther target, thereby ensuring the beam reflection returning to the receiver. Yes.

JP-A-6-230135

  However, in the conventional example disclosed in Patent Document 1 described above, when a series of reflectors such as road markers existing around the vehicle is detected, the reflectors are caused by the presence of other reflectors such as a reflector of the preceding vehicle. There was a problem of misdetecting.

  The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide an object detection apparatus capable of detecting a reflector group existing around a vehicle with high accuracy. Is to provide.

In order to achieve the above object, the present invention provides a light projecting unit that projects light, an image capturing unit that captures surroundings including reflected light from light projected from the light projecting unit, and an image captured by the image capturing unit. A reflected light extraction unit for detecting reflected light from Furthermore, a reflection object group determination unit is provided that determines whether each reflection object is a series of reflection object groups based on the relative luminance of the reflection light detected by the reflection light extraction unit. The reflector group determination unit determines whether or not each reflector is a series of reflector groups based on the position of the reflected light from each reflector in the horizontal direction on the image and the luminance.

  In the object detection device according to the present invention, since it is determined whether or not each reflector is a series of reflectors based on the relative luminance of the reflected light detected by the reflected light extraction unit, it exists around the vehicle. It is possible to detect the reflector group to be detected with high accuracy.

It is a block diagram which shows the structure of the object detection apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing which shows the delineator, cat's eye, and preceding vehicle which exist in the travel path ahead of a vehicle. It is explanatory drawing which shows the reflected light image by the delineator, cat's eye, and preceding vehicle which exist in the travel path ahead of a vehicle. It is a flowchart which shows the process sequence of the object detection apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing which shows the reflected light image by the delineator which exists in the 1st modification of 1st Embodiment, a cat's eye, and a preceding vehicle which exist in the travel path ahead of a vehicle. It is explanatory drawing which shows the delineator and cat's eye which exist in the driving | running road ahead of a vehicle in the case of a 2nd modification of 1st Embodiment, and a driving | running road is curving. It is a block diagram which shows the structure of the object detection apparatus which concerns on 2nd Embodiment of this invention. It is explanatory drawing which shows the optical flow by the delineator, cat's eye, and preceding vehicle which exist in the travel path ahead of a vehicle. It is a flowchart which shows the process sequence of the object detection apparatus which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the structure of the object detection apparatus which concerns on 3rd Embodiment of this invention. It is explanatory drawing which shows the front estimation image and road marker image by the object detection apparatus which concerns on 3rd Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[Description of Configuration of First Embodiment]
FIG. 1 is a block diagram showing the configuration of the object detection apparatus according to the first embodiment of the present invention. As illustrated in FIG. 1, the object detection apparatus 101 according to the first embodiment includes a light projecting unit 11, an imaging unit 12, a reflected light extraction unit 13, and a reflecting object group determination unit 14. And the road marker (reflecting object arrange | positioned continuously) which exists in the road side surrounding a vehicle travel path and the road lane line periphery is detected. Examples of road markers include a delineator (Delineator) and a cat's eye (Cat's eye).

  The reflected light extraction unit 13 and the reflected object group determination unit can be configured as an integrated computer including a central processing unit (CPU), storage means such as a RAM, a ROM, and a hard disk.

  The light projecting unit 11 is provided at an appropriate position in the front visually recognizing position of the vehicle, and projects flashing light that repeatedly turns on and off at a constant period toward the front of the vehicle. When a synchronous detection method described later is adopted, pattern light whose luminance changes periodically is projected.

  The imaging unit 12 is a camera equipped with a CMOS imager, for example, and images the surroundings of the vehicle. The image obtained by imaging includes the reflected light that is projected from the light projecting unit 11 and reflected by the reflecting object X1 such as a road marker existing ahead. In the image obtained by imaging, the horizontal axis direction is the horizontal direction, and the vertical axis direction is the front-rear direction.

  The reflected light extraction unit 13 extracts a region where blinking occurs in the image captured by the imaging unit 12, that is, a region where the luminance changes periodically. Let the image of the extracted area | region be a reflected light image. The reflected light that is repeatedly turned on and off periodically can be regarded as reflected light of light projected from the light projecting unit 11 instead of sunlight or irradiation light from other light sources. It can be estimated that this is a reflector such as a road marker existing in the area. Therefore, by extracting the reflected light image, an image that is a candidate for the road marker can be obtained.

  The reflection object group determination unit 14 determines whether or not each reflection object is a series of reflection object groups based on the relative luminance relationship of the reflected light image of each reflection object extracted by the reflection light extraction unit 13. to decide. Specifically, the reflected light image extracted by the reflected light extraction unit 13 is extracted when the luminance exceeds a predetermined threshold (for example, a luminance that is twice the detection noise level), and the candidate point of the road marker is extracted. To do. That is, a retroreflective material is attached to a road marker such as a delineator or a cat's eye, and the brightness of the reflected light is estimated to be higher than the brightness of the reflected light of other objects. Is extracted as a candidate point for a road marker.

  Furthermore, when one candidate point is determined, the reflector group determination unit 14 selects a region near the candidate point, for example, a region that is 10 times the vertical axis direction and 5 times the horizontal axis direction of the size of the reflected light image. Set this as the neighborhood area. Then, candidate points are extracted in the vicinity region by the same procedure as described above. By repeating this, all candidate points existing in the entire image are extracted. That is, candidate points for road markers are extracted from the entire image.

  For example, as shown in FIG. 2, a series of delineator groups 21 including a plurality of delineators 21 a exist on the left side of the road ahead of the vehicle, and a series of cats eyes 22 including a plurality of cat eyes 22 a on the right side of the vehicle. Is assumed, and further, the preceding vehicle 31 is present in front of the traveling road. Then, in the image captured by the imaging unit 12, the reflected light from each delineator 21a and each cat's eye 22a is extracted by the reflected light extraction unit 13 as light having high luminance (light having a luminance exceeding a predetermined value). The In addition to this, the light reflected by the reflector 32 of the preceding vehicle 31 is extracted as light with high luminance. Therefore, the area | region shown by the code | symbol 21, 22, 32 shown to Fig.3 (a) as a candidate point of a road marker is extracted as a candidate point. The reflector group determination unit 14 performs a process of extracting a series of delineator groups 21 and a series of cat's eye groups 22 based on the position of each candidate point and the relative luminance relationship. A detailed extraction method will be described later.

[Description of Operation of First Embodiment]
Next, the operation of the object detection apparatus according to this embodiment configured as described above will be described with reference to the flowchart shown in FIG. First, in step S11, the light projecting unit 11 projects flashing light toward the front of the vehicle.

  In step S12, the imaging unit 12 images the front of the vehicle. The image obtained by imaging includes the reflected light that is projected from the light projecting unit 11 and reflected by the reflector X1 (see FIG. 1) or the like existing in front of the vehicle. By imaging the front of the vehicle with the imaging unit 12, for example, an image as shown in FIG. 2 is obtained. In this image, there are a delineator group 21 composed of a plurality of delineators 21a, a cat's eye group 22 composed of a plurality of cat's eyes 22a, and a reflector 32 of a preceding vehicle 31.

  In step S <b> 13, the reflected light extraction unit 13 extracts reflected light from the image captured by the imaging unit 12 and sets this as a reflected light image. As described above, since the light projected from the light projecting unit 11 is blinking, the luminance periodically changes. Therefore, the reflected light image of the light projected from the light projecting unit 11 can be extracted by calculating the difference between the image when turned on and the image when turned off.

  In step S <b> 14, the reflector group determination unit 14 extracts the one whose luminance exceeds a predetermined threshold based on each reflected light image extracted in the process of step S <b> 13. That is, the delineator 21a and the cat's eye 22a existing in front of the vehicle are provided with retroreflective materials, and the light reflected by these retroreflective materials is expected to have high luminance. Accordingly, it is determined that an area having a luminance exceeding a preset threshold is a candidate point for reflected light from a road marker such as the delineator 21a or the cat's eye 22a. By this process, for example, an image from which a plurality of candidate points are extracted is obtained as shown in FIG.

  In step S15, the reflector group determination unit 14 plots the position in the vertical axis direction (position in the vehicle front-rear direction) on the image of each candidate point and the luminance (reception intensity) in association with each other on a plane. As a result, as shown in FIG. 3B, the reflected light images from the delineators 21a, the cat's eyes 22a, and the reflectors 32 are respectively point 21b (black circle), point 22b (+ mark), and point 32b (*). Are marked). Each point is connected with a smooth curve. As a result, a curve Q1 corresponding to the delineator group 21 and a curve Q2 corresponding to the cat's eye group 22 are obtained. At this time, the starting points of the curves Q1 and Q2 are defined as vanishing points of the traveling road (points where the left road side and the right road side intersect on the image). Further, the point 32b by the reflector 32 is not connected to other points by a smooth curve, and therefore does not belong to any of the curves Q1 and Q2.

  Thereafter, in step S16, the reflector group determination unit 14 determines whether or not the luminance changes of the curves Q1 and Q2 shown in FIG. 3B have a quadratic function characteristic. That is, since the luminance is inversely proportional to the square of the distance, the reflected light from each delineator 21a provided at substantially equal intervals on the road side is estimated to have a quadratic function characteristic. By verifying this, It is determined that the road marker is a delineator group 21 or the like in which reflective objects are continuously arranged. The verification of whether or not it has the characteristics of a quadratic function can determine whether or not it is a series of reflectors based on the luminance ratio of adjacent candidate points (reflected light). As an example, the determination can be made based on whether or not the luminance ratio of the candidate points adjacent to each other is about ½.

  If it is determined in step S17 that the characteristic has a quadratic function, the reflector group determination unit 14 determines that the candidate points corresponding to the respective points 21b shown in FIG. It is estimated that the image is a reflected light image by the delineator 21a shown in FIG. Similarly, it is estimated that the candidate point corresponding to each point 22b shown in FIG. 3B is a reflected light image by the cat's eye 22a shown in FIG. Accordingly, the object present on the left road side in the front image shown in FIG. 2 is recognized as a series of delineators 21. Further, the object present on the right side of the road is recognized as a series of cat's eye groups 22.

  In this way, the reflected light from the delineator group 21 and the reflected light from the cat's eye group 22 can be selected and extracted from the plurality of candidate points shown in FIG. 3A, and the delineator and cat's eye existing in the front of the vehicle can be extracted. Or the like can be detected. Moreover, the reflected light by the reflector 32 of the preceding vehicle 31 can be excluded.

  Thus, in the object detection apparatus 101 according to the first embodiment, the flashing light that repeatedly turns on and off at a constant cycle is projected from the light projecting unit 11 around the vehicle, and the object that exists around the vehicle. The reflected light reflected at is picked up by the image pickup unit 12. And the candidate point of a road marker is extracted from the reflected light imaged by the imaging part 12. FIG. Further, the relationship between the position of each candidate point and the luminance is plotted on a plane, and when each point is connected by a smooth curve, it is determined whether or not it has a quadratic function characteristic (relative luminance relationship). In the case of having the next function characteristic, this candidate point is determined to be a series of road marker groups existing around the vehicle. Therefore, road markers such as the delineator 21a and the cat's eye 22a existing around the host vehicle can be detected with high accuracy.

  Further, even when a plurality of delineators 21a and a plurality of cat's eyes 22a are mixed or when light reflected by the reflector 32 of the preceding vehicle 31 is mixed, these are surely selected and classified into the delineator 21a and the cat's eyes 22a. And false detection can be reduced or eliminated altogether.

  Furthermore, since it is verified whether each candidate point is a series of reflector groups based on the luminance ratio of the candidate points (reflected light), it is possible to detect a road marker group with higher accuracy.

  In the embodiment described above, a smooth curve is connected starting from the vanishing point of the traveling road ahead of the vehicle, but without using the vanishing point, the candidate point existing at the top of the vertical axis is used as the starting point. You may make it connect a smooth curve.

  Further, as shown in FIG. 3 (c), when the luminance with respect to the position in the horizontal direction of the image of each candidate point is plotted and each point 21b is connected by a smooth curve Q4, these are represented by a series of road markers. It is also possible to determine that there is.

[Description of First Modification of First Embodiment]
Further, as shown in FIG. 5, the positions of a series of road marker candidate points in the image are detected, and the height of the retroreflective material provided on the road marker is determined based on the vertical position coordinates of the detected positions. Can be recognized. That is, the position 21P of the lowest point of the delineator group 21 shown in FIG. 5A and the position 22P of the lowest point of the cat's eye group 22 are vertically arranged on the image as shown in FIG. 5B. There is a big difference. Therefore, it is determined that the candidate point indicated by reference numeral 21 is the retroreflective material provided at a position higher than the road surface, and the candidate point indicated by reference numeral 22 is the one provided by the retroreflective material at a position lower than the road surface. It can be judged.

  From the above, the candidate point indicated by reference numeral 21 is reflected light by a delineator provided with a retroreflective material at a high position from the road surface, and the candidate point indicated by reference numeral 22 is reflected light by a cat's eye provided on the road surface. Can be recognized. Thus, by detecting the vertical position of the lowest point of each candidate point, it becomes possible to recognize the height of the road marker from the road surface.

[Description of Second Modification of First Embodiment]
In the first embodiment described above, it is assumed that the vehicle is traveling on a substantially straight traveling path, and the position of each candidate point with respect to the position of each candidate point in the vertical axis direction (front-rear direction of the vehicle). The luminance was plotted to create curves Q1 and Q2 shown in FIG. However, when the traveling road ahead of the vehicle is curved, there are cases where the characteristic curve cannot be obtained when the luminance is plotted against the position in the vertical axis direction. In the second modification, a method for creating the characteristic polarity in such a case will be shown.

  For example, as shown in FIG. 6, a case is assumed in which the traveling road ahead of the vehicle is greatly curved in the right direction. A delineator group 21 exists on the left side of the traveling path, and a cat's eye group 22 exists on the right side of the traveling path. In such a case, each candidate point may overlap in the vertical axis direction of the image. Therefore, when the luminance of each candidate point is plotted as shown in FIG. Smooth curves may not be created.

  In the first modification, the luminance of each candidate point is plotted in the horizontal axis direction of the image. As a result, a curve Q3 as shown in FIG. 6B can be generated. When the curve Q3 becomes a smooth curve, it is determined that each candidate point is a series of delineator groups 21. Further, the presence of the cat's eye group 22 existing on the right side of the road can be similarly recognized by plotting the luminance of each candidate point in the horizontal axis direction of the image.

  Further, since the candidate point of the delineator 21a and the candidate point of the cat's eye 22a have different luminances, it is possible to select and recognize each candidate point even when the candidate points are mixed.

  Thus, in the second modification, even when a plurality of points indicating luminance in the vertical axis direction of the image cannot be plotted, such as when the vehicle travels on a curved road, the luminance is plotted in the horizontal axis direction of the image. Thus, it is possible to detect with high accuracy whether or not it is a series of road markers.

[Description of Third Modification of First Embodiment]
Next, a third modification of the first embodiment described above will be described. In the first embodiment described above, flashing light that repeatedly turns on and off at a constant cycle is projected from the light projecting unit 11, and the difference between the image at the time of lighting and the image at the time of light off that is captured by the imaging unit 12 is calculated. Thus, an example in which reflected light reflected by a reflecting object such as a road marker is extracted has been described.

On the other hand, in the third modification, a synchronous detection method is adopted to detect the light projected and the light captured in synchronization. Specifically, a signal transmitted from the light projecting unit 11 is S (t), and this is modulated with sin (ωt). Further, the image data captured by the imaging unit 12 is multiplied by sin (ωt). Then, the signal S (t) can be extracted from the relationship of the following formula (1) by removing the high frequency component with a low-pass filter.
S (t) * sin (ωt) * sin (ωt) = S (t) * (1-cos (2ωt)) / 2 (1)
Since the synchronous detection method is a well-known technique, detailed description thereof is omitted.

  Thus, in the object detection device according to the third modification, the light projected from the light projecting unit 11 and the reflected light included in the image captured by the image capturing unit 12 are synchronized using the synchronous detection method. Since it can be detected, only the reflected light can be extracted with high accuracy by a reflecting object such as a road marker that is projected from the light projecting unit 11, and more accurate object detection can be performed.

[Description of Second Embodiment]
Next, a second embodiment of the present invention will be described. FIG. 7 is a block diagram illustrating a configuration of the object detection device according to the second embodiment. As shown in FIG. 7, the object detection apparatus 102 according to the second embodiment is different from FIG. 1 in that a reflected light movement amount measurement unit 15 is provided. Since the other configuration is the same as that of FIG. 1, the same reference numerals are given and the description of the configuration is omitted.

  The reflected light movement amount measurement unit 15 (optical flow calculation unit) calculates an optical flow of reflected light included in an image captured at a predetermined cycle, and moves the reflected light for each predetermined cycle based on the optical flow. Calculate the amount. Then, the reflector group determination unit 14 determines whether the reflected light is a series of road markers based on the amount of movement of each reflected light and the luminance.

  Hereinafter, the principle of detecting the reflected light by the road marker will be described with reference to FIG. FIG. 8A shows candidate points obtained from the image captured at the previous imaging, and candidate points obtained from the image captured at the current imaging, and the movement amount of each candidate point as an optical flow. Show. From FIG. 8 (a), it is understood that the moving amount is smaller and the optical flow length is shorter as the candidate point exists further forward.

  Then, as shown in FIG. 8B, the position in the vertical direction of the image and the flow length of the optical flow are plotted. Further, as shown in FIG. 8C, each candidate point is plotted on coordinates where the horizontal axis is the flow length of each candidate point and the vertical axis is the luminance of each candidate point. If the plotted candidate points are present in a straight line, it is determined that each candidate point is reflected light from a series of road markers.

  Hereinafter, the processing procedure of 2nd Embodiment is demonstrated with reference to the flowchart shown in FIG. First, in step S31, the light projecting unit 11 projects flashing light toward the front of the vehicle.

  In step S32, the imaging unit 12 images the front of the vehicle. The image obtained by imaging includes the reflected light that is projected from the light projecting unit 11 and reflected by the reflector X1 (see FIG. 1) or the like existing in front of the vehicle.

  In step S <b> 33, the reflected light extraction unit 13 extracts reflected light from the image captured by the imaging unit 12 and sets this as a reflected light image. As described above, since the light projected from the light projecting unit 11 is blinking, the luminance periodically changes. Therefore, the reflected light image of the light projected from the light projecting unit 11 can be extracted by calculating the difference between the image when turned on and the image when turned off.

  In step S <b> 34, the reflector group determination unit 14 extracts an object whose luminance exceeds a predetermined threshold based on each reflected light image extracted in the process of step S <b> 33. That is, the delineator 21a and the cat's eye 22a existing in front of the vehicle are provided with retroreflective materials, and the light reflected by these retroreflective materials is expected to have high luminance. Accordingly, it is determined that an area having a luminance exceeding a preset threshold is a candidate point for reflected light from a road marker such as the delineator 21a or the cat's eye 22a.

  In step S35, the optical flow of each candidate point is calculated based on the position of the candidate point at the previous imaging time and the position of the candidate point at the current imaging time to generate the image shown in FIG. Thereafter, the movement amount (flow length) of each candidate point is obtained. As a result, as shown in FIG. 8B, movement amount data corresponding to the position in the vertical axis direction of the image is obtained.

  In step S36, as shown in FIG. 8C, the relationship between the movement amount and the luminance displacement (difference between the previous luminance and the current luminance) is plotted on the coordinates. Then, it is determined whether each candidate point has a linear characteristic (whether it is linear).

  In step S37, if the points indicating the amount of movement and the brightness of each candidate point have linear characteristics, these candidate points can be recognized as a series of road marker groups. That is, as shown in FIG. 8A, it is recognized that the candidate points of each delineator 21 a are a series of delineator groups 21. Further, the reflected light of each cat's eye 22a and the reflector 32 of the preceding vehicle 21 is excluded because it does not exist on the straight line shown in FIG. 8C. Although not shown, the candidate points of the cat's eye 22a have a linear characteristic different from that in FIG. 8C, and the cat's eye group 22 can be recognized based on this linear characteristic.

  In this way, in the second embodiment, the optical flow of the candidate points is calculated from each image captured in time series to obtain the movement amount of each candidate point, and the movement amount and luminance displacement of each candidate point are calculated. Whether each candidate point is a series of road markers is determined based on whether the relationship has linearity. Therefore, as in the first embodiment, road markers such as the delineator 21a and the cat's eye 22a existing around the host vehicle can be detected with high accuracy.

[Description of Third Embodiment]
Next, a third embodiment of the present invention will be described. FIG. 10 is a block diagram illustrating a configuration of the object detection device according to the third embodiment. As shown in FIG. 10, the object detection apparatus 103 according to the third embodiment includes a travel boundary detection unit 16, an imaging unit layout information database 17, a map matching unit 18, and a map database 19, as compared with FIG. 1. Is different. Since the other configuration is the same as that of FIG. 1, the same reference numerals are given and the description of the configuration is omitted.

  The travel boundary detection unit 16 extracts the travel boundary line of the travel path ahead of the vehicle based on the position of the road marker group included in the image captured by the imaging unit 12.

  The imaging unit layout information database 17 stores information on the position and angle of the imaging unit 12 attached to the vehicle (layout information). The map database 19 (map data storage unit) stores map data of a travel route on which the vehicle travels.

  The map matching unit 18 includes map data stored in the map database 19, image capturing unit layout (position and angle information) stored in the image capturing unit layout information database 17, and the current position of the vehicle (not shown GPS). Based on the position information specified by the device or the like, a traveling path ahead of the vehicle is predicted, and an estimated image is created. That is, the map collation unit 18 has a function as a position detection unit that detects the attachment state of the imaging unit 12. Further, the map collation unit 18 collates the image of the travel boundary (image of the road marker group) detected by the travel boundary detection unit 16. Then, both images are compared while changing the current candidate position of the host vehicle on the map based on the imaging unit layout, and an estimated image with a minimum norm is selected. And the imaging part layout from which this estimated image is obtained, and the present candidate position of the own vehicle are recognized.

  For example, when an estimated image of the travel boundary as shown in FIG. 11A is obtained, and the travel boundary image detected by the travel boundary detection unit 16 is an image as shown in FIG. Selects an imaging unit layout that minimizes these differences.

  Then, an estimated image is created based on the selected image capturing unit layout information, map data, and current vehicle position data, and is used for estimating a road marker with reference to the estimated image.

  That is, whether or not the extracted candidate points are a series of road markers by determining whether the position of the road marker extracted by the reflector group determination unit 14 matches the traveling boundary included in the estimated image. Determine whether. At this time, candidate points existing at positions far away from the travel boundary can be excluded from the candidate road markers.

  Thus, in the object detection apparatus 103 according to the third embodiment, the traveling boundary ahead of the vehicle is estimated based on the layout information of the imaging unit 12, the current position of the vehicle, and the map data, and the estimated traveling boundary is determined. A series of road marker groups are extracted by excluding candidate points away from. Therefore, it is possible to prevent the retroreflecting material present at a position away from the travel boundary from being erroneously detected as a road marker, and to detect the road marker with higher accuracy.

  Although the object detection device of the present invention has been described based on the illustrated embodiment, the present invention is not limited to this, and the configuration of each part is replaced with an arbitrary configuration having the same function. Can do.

DESCRIPTION OF SYMBOLS 11 Light projection part 12 Imaging part 13 Reflected light extraction part 14 Reflected object group judgment part 15 Reflected light movement amount measurement part 16 Traveling boundary detection part 17 Imaging part layout information database 18 Map collation part 19 Map database 21 Deliniator group 21a Delineator 22 Cats Eye group 22a Cat's eye 31 Leading vehicle 32 Reflector 101, 102, 103 Object detection device X1 Reflecting object

Claims (5)

An object detection device for detecting a reflector group composed of a plurality of reflectors arranged around a vehicle travel path,
A light projecting unit that projects light around the vehicle;
An imaging unit that captures an image of the surroundings including the reflected light reflected by each of the reflectors, the light projected from the light projecting unit;
A reflected light extraction unit that detects the reflected light from an image captured by the imaging unit;
A reflector group determination unit that determines whether or not each reflector is a series of reflector groups based on the relative luminance of the reflected light from each reflector detected by the reflected light extraction unit ;
The reflector group determination unit determines whether or not each reflector is a series of reflector groups based on the horizontal position on the image of the reflected light from each reflector and the luminance. An object detection device characterized by. The light projecting unit projects light whose luminance changes periodically,
The object detection apparatus according to claim 1, wherein the reflected light extraction unit extracts reflected light synchronized with the change in luminance from an image captured by the imaging unit. The reflector group determination unit determines whether each reflected light is a series of reflector groups based on the position of each reflected light and the luminance ratio of each reflected light on the image captured by the imaging unit. The object detection device according to claim 1, wherein: An optical flow calculation unit that calculates an optical flow of each reflecting object based on the reflected light from the reflecting object;
The said reflector group judgment part judges whether each said reflector is a series of reflector groups based on the magnitude | size of the said optical flow, The any one of Claims 1-3 characterized by the above-mentioned. The object detection device according to item. A map data storage unit that stores map data, a GPS device that identifies the position of the vehicle, and a position detection unit that detects a mounting state of the imaging unit;
The reflector group determination unit identifies a position of the vehicle on the map data, and further predicts a traveling path existing in front of the vehicle based on the mounting position of the imaging unit detected by the position detection unit. And
Based on the predicted traveling path and the position on the image of the reflected light by the reflector, it is determined whether or not each reflector is a series of reflectors.
The object detection device according to claim 1, wherein

Images