JP7334479B2 - Object detection system and object detection method - Google Patents

Description

The present invention relates to an object detection system and an object detection method.

2. Description of the Related Art A technique for detecting a specific object such as a crosswalk using an image captured by an in-vehicle camera is known (see Patent Document 1, for example). The technology described in Patent Literature 1 detects a crosswalk by performing pattern matching on the white line portion forming the crosswalk and the black portion on the road in the captured image. The technique described in Patent Document 2 easily detects a curb surface perpendicular to the road surface by recognizing a plurality of planes formed by a plurality of edge lines in the captured image. The roadside object detection device described in Patent Document 3 uses three-dimensional information in the vicinity of the imaging unit in the captured image to improve detection accuracy of roadside objects far from the imaging unit.

JP 2011-192071 A Japanese Patent Application Laid-Open No. 2007-18037 Japanese Unexamined Patent Application Publication No. 2014-2608

The techniques described in Patent Documents 1 to 3 are based on the premise that the image patterns of the pedestrian crossing and curb stones to be detected are clearly captured in the captured image. Therefore, in the system using the techniques described in Patent Documents 1 to 3, there is a possibility that the detection target cannot be detected because the image pattern of the detection target does not appear clearly in the captured image at night or in rainy weather. Such a problem becomes more conspicuous as the distance from the imaging unit increases. Therefore, even at night or in rainy weather, there has been a problem of detecting a detection target regardless of whether it is near or far from the imaging unit.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and it is an object of the present invention to clearly detect a detection target that is the source of an unidentifiable image pattern from a captured image.

The present invention has been made to solve the above-described problems, and can be implemented as the following modes. An object detection system, comprising: a light emitting device arranged on or near an object provided on a road, the light emitting device being capable of modulating light emission conditions over time; and an imaging unit that captures an image of the light-emitting device located in the vicinity of the target, and an object that is pre-associated with the light-emitting condition of the light-emitting device by analyzing changes over time in the image captured by the imaging unit. a detector, wherein the light-emitting device is placed on a curb of a road and emits light with wavelengths having different directivities depending on the direction of light emission; and the object detector detects the position of the curb as an object. and, when the captured image includes a plurality of the light emitting devices, the shape of the road is detected using the positions of the curbs corresponding to the light emitting devices and the directivity of the light emitting devices. , object detection system. An object detection system, comprising: a light emitting device arranged on or near an object provided on a road, the light emitting device being capable of modulating light emission conditions over time; and an imaging unit that captures an image of the light-emitting device located in the vicinity of the target, and an object that is pre-associated with the light-emitting condition of the light-emitting device by analyzing changes over time in the image captured by the imaging unit. a detection unit, wherein the light-emitting devices are arranged at least two locations on a crosswalk on a road, and the object detection unit detects the position of the crosswalk as an object, An object detection system that uses the distance between devices to calculate the distance from the vehicle to the pedestrian crossing. In addition, the present invention can also be implemented as the following modes.

(1) According to one aspect of the present invention, an object detection system is provided. This object detection system comprises a light-emitting device arranged on or near an object provided on a road, the light-emitting device being capable of temporally modulating light emission conditions, and a light-emitting device mounted on a vehicle. an imaging unit that captures an image of the light emitting device located in the vicinity; and an object detection that detects an object that is pre-associated with the light emission condition of the light emitting device by analyzing changes over time in the image captured by the imaging unit. and

According to this configuration, the subject detection section analyzes changes over time in the captured image of the light emitting device captured by the imaging section. The object detection unit detects objects pre-associated with the light emission conditions of the light emitting device. As a result, for example, a white road surface pattern such as a stop line and a crosswalk provided on the road is detected by the object detection unit as the light emission condition of the light emitting device, not as an image pattern in the captured image. As a result, the object detection unit can clearly detect an image pattern of an object located away from the vehicle and an object that is difficult to extract from the captured image as the light emission condition.

(2) In the object detection system of the aspect described above, the light emitting device is arranged on a stop line of a road, and the object detection unit includes the stop line as the object and the vicinity of the stop line as the object. at least one of the position of a crosswalk placed in the
According to this configuration, at least one of the stop line and the pedestrian crossing arranged near the stop line is detected by the object detection unit as the object associated with the light emitting device. This allows the object detection unit to clearly detect the stop line where the vehicle should stop and the crosswalk where pedestrians may be present. As a result, vehicles equipped with object detection systems can safely travel on roads.

(3) In the object detection system of the aspect described above, the light emitting device may be arranged on a curb of a road, and the object detection section may detect the position of the curb as the object.
According to this configuration, the object detection unit can clearly detect the curb that separates the road from the portion other than the road, so that the driver of the vehicle can travel safely on the road.

(4) In the object detection system of the aspect described above, when the captured image includes a plurality of the light emitting devices, the object detection unit detects the shape of the road from the position of each of the curbs corresponding to each of the light emitting devices. may be detected.
According to this configuration, the object detection unit detects the shape of the road by detecting the position of the light emitting device on the road. As a result, the driver of the vehicle can recognize the shape of the road on which the vehicle is to travel, so that the driver can travel on the road more safely.

(5) In the object detection system of the aspect described above, the light-emitting device emits light having directivity, and the target detection unit detects the positions of the curbs corresponding to the light-emitting devices and the positions of the respective light-emitting devices. Directivity may be used to detect the shape of the road.
According to this configuration, the object detection unit uses the light emission pattern of the light emitting device to detect the orientation of the light emitting device by using directional light emission in addition to the position of the light emitting device on the road. This allows the object detection unit to more clearly detect the shape of the road.

(6) In the object detection system of the aspect described above, the light emitting device may be arranged on a road crosswalk, and the object detection unit may detect the position of the crosswalk as the object.
A vehicle that employs this configuration can detect a target crosswalk to stop, so it can travel safely on the road.

(7) In the object detection system of the aspect described above, the light-emitting devices are arranged in at least two locations on the crosswalk, and the object detection unit uses the distances between the light-emitting devices in the captured image to detect the A distance from the vehicle to the pedestrian crossing may be calculated.
According to this configuration, the distance from the vehicle adopting this configuration to the pedestrian crossing can be calculated using the distance between the two light emitting devices. As a result, the vehicle can measure the distance to the crosswalk to which attention should be paid and control the operation such as deceleration, so that the vehicle can travel more safely.

(8) In the object detection system of the above aspect, the light emission condition is a pulsed light emission pattern in which the light emission of the light emitting device is repeatedly turned on and off over time, and the light emission pattern is the wavelength of the pulse. , at least one of the emission intensity of the pulse and the width and period of the pulse may be varied.
According to this configuration, by changing the combination of pulse wavelengths, etc., it is possible to realize various light emission patterns that can be easily distinguished and detected by the target detection unit.

(9) In the object detection system of the aspect described above, the light emission pattern of the light emitting device is set according to the type of the object with which the light emitting device is associated, and the object detection unit detects the light emitting device in the captured image. Different objects may be distinguished and detected by analyzing the synchronization of the light emission patterns of .
According to this configuration, the synchronization of the light emission pattern of the light emitting device differs for each type of object. Accordingly, the object detection unit can more clearly distinguish and detect the object by analyzing the imaged images taken in chronological order and divided into specific intervals using synchronization.

(10) According to another aspect of the invention, there is provided an object detection method. According to this object detection method, an object provided on a road or a light emitting device placed near the object and capable of temporally modulating the light emission condition is imaged by an imaging unit mounted on a vehicle. an analysis step of analyzing temporal changes in the captured image of the imaging unit; and a detection step of detecting an object pre-associated with the light emission conditions of the light emitting device using the analysis result. .

It should be noted that the present invention can be implemented in various aspects, including, for example, an object detection system, a control device, a vehicle equipped with these devices, an object detection method, and a method for executing these devices and methods. It can be implemented in the form of a computer program, a server device for distributing this computer program, a non-temporary storage medium storing the computer program, or the like.

1 is a schematic diagram of an object detection system as a first embodiment of the present invention; FIG. It is explanatory drawing which shows an example of the captured image of a vehicle-mounted camera. 4 is a flowchart of an object detection method according to the present embodiment; It is a schematic diagram of an object detection system in a 2nd embodiment. FIG. 4 is an explanatory diagram for calculating the distance from the vehicle to the pedestrian crossing; It is a schematic diagram of a light-emitting device in a third embodiment. 1 is a schematic diagram of a captured image including a light emitting device placed on a curb of a road; FIG. It is explanatory drawing which shows the captured image of the vehicle-mounted camera in 4th Embodiment.

<First embodiment>
FIG. 1 is a schematic diagram of an object detection system 100 as a first embodiment of the invention. The object detection system 100 is a system that detects an object pre-associated with a light-emitting device imaged by the vehicle-mounted camera 11 by analyzing changes over time in images captured by the vehicle-mounted camera 11. As shown in FIG. 1, the object detection system 100 includes a light-emitting device 51 placed on the road stop line HL and a vehicle 10 traveling on the road RD.

The light emitting device 51 of this embodiment is a light emitting diode (LED). The light emission conditions of the light emitting device 51 are modulated over time. The light emission condition of the light emitting device 51 is, for example, a light emission pattern of pulses in which light emission is repeatedly turned on and off over time. In the present embodiment, at least one of the pulse wavelength, the pulse emission intensity, and the pulse width and cycle is different for each emission pattern.

As shown in FIG. 1, a vehicle 10 includes an on-vehicle camera (imaging unit) 11 that captures an image of an external scene including light emitting devices 51 positioned around the vehicle 10, and a control device 20 that controls each unit of the vehicle 10. ing. The vehicle-mounted camera 11 is arranged near a rearview mirror in the vehicle 10 . The in-vehicle camera 11 images the front of the vehicle 10 as the surroundings of the vehicle 10 .

The control device 20 includes a CPU (Central Processing Unit) 30, a ROM (Read Only Memory) 21, a RAM (Random Access Memory) 22, and a storage section 40 for storing various data. The CPU 30 executes various programs stored in the ROM 21 by developing them in the RAM 22 .

The storage unit 40 is configured by a hard disk drive (HDD) or the like. The storage unit 40 includes a map database (map DB) 41 that stores map information, and an object database (object DB) 42 that stores objects associated in advance with the light emission conditions of the light emitting device 51 . , is equipped with In the target object DB 42 of the present embodiment, the correspondence relationship between the light emission pattern of the light emitting device 51 captured by the vehicle-mounted camera 11, the stop line HL as the target object, and the position of the pedestrian crossing CW arranged near the stop line HL. is stored.

The CPU 30 includes a communication unit 31 that wirelessly communicates information with other devices, an operation control unit 32 that controls the operation of the vehicle 10 in accordance with a driver's operation input such as a steering wheel (not shown), and an in-vehicle camera 11. It functions as a detection unit 33 that detects an object such as the stop line HL by analyzing temporal changes in captured images, and an image processing unit 34 . The communication unit 31 receives various objects associated with light emission patterns of various light emitting devices and map information transmitted from other devices.

The image processing unit 34 performs various types of processing on the captured image captured by the vehicle-mounted camera 11 . The detection unit 33 identifies the positions of the various light emitting devices with respect to the vehicle 10 from the positions of the various light emitting devices in the images captured by the vehicle-mounted camera 11 . Note that the detection unit 33 and the image processing unit 34 in this embodiment function as a target detection unit.

FIG. 2 is an explanatory diagram showing an example of an image IM1 captured by the in-vehicle camera 11. As shown in FIG. A captured image IM1 shown in FIG. 2 is an image obtained by combining the captured images of the vehicle-mounted camera 11 by the image processing unit 34 . The captured image IM1 includes the stop line HL and the crosswalk CW placed on the road RD, and the light emitting device 51 placed on the stop line HL. As shown in captured image IM1, vehicle 10 is traveling on road RD immediately before the intersection (T-junction). The crosswalk CW is located farther from the vehicle 10 than the stop line.

The light emission pattern of the light emitting device 51 is associated with the stop line HL and the position of the crosswalk CW arranged near the stop line HL in the object DB 42 . Therefore, when detecting unit 33 detects a temporal change in the light emission pattern of light emitting device 51 in captured image IM1, the detected light emission pattern is checked against the correspondence stored in object DB 42 . When the same light emission pattern as the detected light emission pattern is stored in the object DB 42, the detection unit 33 detects the positions of the stop line HL and the crosswalk CW as objects associated with the light emission pattern.

FIG. 3 is a flowchart of the object detection method (hereinafter also simply referred to as "detection flow") in this embodiment. In the detection flow shown in FIG. 3 , first, the vehicle-mounted camera 11 performs an imaging process of imaging the surroundings of the vehicle 10 . (Step S1). Next, the detection unit 33 detects the light emitting device 51 from the captured image IM1 (step S2). When the light emitting device 51 is detected from the captured image IM1 (step S2: YES), the detection unit 33 analyzes the temporal change in the light emission condition of the detected light emitting device 51 (step S3).

As a result of the analysis, the detection unit 33 determines whether or not the same light emission pattern as the detected light emission pattern is stored in the target object DB 42 (step S4). When the same light emission pattern as the detected pattern is not stored in the target object DB 42 (step S4: NO), the processes after step S2 are performed. When the same light emission pattern as the detected light emission pattern is stored in the target object DB 42 (step S4: YES), the detection unit 33 detects the detected light emitting device 51 as the target object associated in advance. A detection step of detecting the positions of the stop line HL and the crosswalk CW is performed (step S5), and the detection flow ends. Note that the processing from step S2 to step S4 functions as an analysis step.

As described above, in the object detection system 100 of the present embodiment, the detection unit 33 analyzes changes over time in the captured image of the light emitting device 51 captured by the in-vehicle camera 11 . Thereby, the detection unit 33 detects the stop line HL pre-associated with the light emission condition of the light emitting device 51 . Therefore, the object detection system 100 of the present embodiment detects a white road surface pattern such as the stop line HL and the pedestrian crossing CW provided on the road RD using the light emission of the light emitting device 51 instead of the image in the captured image IM1. It can be detected as a condition. As a result, the detection unit 33 can clearly detect an image pattern of an object located far from the vehicle 10 and an object that is difficult to extract from the captured image IM1 as the light emission condition.

Further, the light emitting device 51 in this embodiment is arranged on the stop line HL. The detection unit 33 detects the stop line HL and the crosswalk CW located near the stop line HL as objects associated with the light emitting device 51 . Therefore, the object detection system 100 of this embodiment clearly detects the stop line HL where the vehicle 10 should stop and the crosswalk CW where pedestrians may be present. Accordingly, the vehicle 10 equipped with the object detection system 100 can safely travel on the road RD.

Further, the light emission condition of the light emitting device 51 of the present embodiment is a pulsed light emission pattern in which light emission is repeatedly turned on and off over time. The emission pattern differs in at least one of the pulse wavelength, the pulse emission intensity, and the pulse width and period. Therefore, the various light-emitting devices of the present embodiment can realize various light-emission patterns that can be easily distinguished and detected by the detection unit 33 by changing combinations of pulse wavelengths and the like.

<Modified Example of First Embodiment>
The present invention is not limited to the above-described embodiments, and can be implemented in various aspects without departing from the scope of the invention. For example, the following modifications are possible.

The position where the light emitting device 51 is arranged may not be on the stop line HL, and may be on any position on the stop line HL. For example, two light emitting devices 51 may be arranged at both ends on the stop line HL. The number of light emitting devices 51 may be two or more, and light emission patterns and LED sizes of various light emitting devices may be different.

The light emission pattern of each light emitting device may be set according to the type of object (stop line HL, pedestrian crossing CW, curbstone, etc.) with which each light emitting device is associated. Further, the detection unit 33 may detect objects of different types by distinguishing between them by analyzing the synchronization of light emission patterns set for each type of object. In this modification, the synchronization of the light emission pattern of the light emitting device differs for each type of object. Therefore, the detection unit 33 can more clearly distinguish and detect the target object by analyzing the captured images IM1 captured in time series and divided into specific intervals using synchronization.

The control device 20 may not necessarily include the map DB 41 and may not function as the communication section 31 and the operation control section 32 . For example, the control device 20 is not mounted on the vehicle 10 but is provided on the cloud or in another device capable of communication, and uses the captured image IM1 transmitted from the communication unit 31 of the vehicle 10 to transmit the target object to the vehicle 10. information may be sent.

<Second embodiment>
FIG. 4 is a schematic diagram of an object detection system 100a in the second embodiment. Compared to the object detection system 100 of the first embodiment, the object detection system 100a of the second embodiment differs from the object detection system 100 of the first embodiment in the positions where the light emitting devices 52 and 53 are arranged. Different processing. Therefore, in the second embodiment, a description of the same configuration and the same control as those of the object detection system 100 of the first embodiment will be omitted, and a different configuration and control will be described.

As shown in FIG. 4, the object detection system 100a includes a light emitting device 52 arranged at one end on the pedestrian crossing CW, a light emitting device 53 arranged in the pedestrian crossing CW, and a vehicle 10a. . The light emitting devices 52 and 53 are the same devices as the light emitting device 51 of the first embodiment. The light-emitting device 53 is arranged near the intersection of the crosswalk CW and the longitudinal extension of the center line CL representing the lane width of the vehicle 10a. Therefore, the width between the light emitting devices 52 and 53 corresponds to the width of the running lane of the vehicle 10a. In other words, the light emitting devices 52 and 53 are arranged at two locations on the pedestrian crossing CW. The position of the pedestrian crossing CW and the distance between the light emitting devices 52 and 53, which are associated in advance with the light emitting patterns of the light emitting devices 52 and 53, are stored in the object DB 42a of the control device 20a.

The detection unit 33a calculates the distance from the vehicle 10 to the crosswalk CW using the change in the number of pixels between the light emitting devices 52 and 53 in the captured image as the vehicle 10 travels. FIG. 5 is an explanatory diagram for calculating the distance from the vehicle 10 to the crosswalk CW. The distance W between the light emitting devices 52 and 53, the distance Z from the vehicle 10 to the crosswalk CW, the focal length f, and the number of pixels B as the length between the light emitting devices 52 and 53 in the captured image are shown in FIG. Defined. Based on these definitions, the detection unit 33a calculates the distance Z from the vehicle 10 to the pedestrian crossing CW by solving the following equation (1).
Z=W.f/B (1)

As described above, the light emitting devices 52 and 53 in the second embodiment are arranged at two locations on the pedestrian crossing CW. The detection unit 33a calculates the distance from the vehicle 10 to the pedestrian crossing CW using the number of pixels as the distance between the light emitting devices 52 and 53 in the captured image. Therefore, the vehicle 10a equipped with the object detection system 100a of the second embodiment can measure the distance to the crosswalk CW to be careful and control the operation such as deceleration, so that the vehicle 10a can travel more safely.

Further, the detection unit 33a in the second embodiment detects the position of the crosswalk CW arranged on the road as the object. Therefore, the vehicle 10a that employs the object detection system 100a of the second embodiment can safely travel on roads.

<Modification of Second Embodiment>
Although the light-emitting device 53 is arranged near the center of the crosswalk CW, the arrangement and number of the light-emitting devices 52 and 53 associated with the crosswalk CW can be modified in various ways. For example, the light emitting devices 52 and 53 may be arranged at both ends of the crosswalk CW. Also, four light emitting devices may be arranged at the four corners defining the area of the crosswalk CW. Although the detection unit 33a uses the focal length f to calculate the distance Z from the vehicle 10 to the pedestrian crossing CW, the distance Z may be calculated using a different method. For example, the detection unit 33a may calculate the distance Z using the vehicle speed and the temporal change in the number of pixels B representing the distance between the light emitting devices 52 and 53 in the captured image.

<Third Embodiment>
FIG. 6 is a schematic diagram of the light emitting device 54 in the third embodiment. In the third embodiment, the configurations of the light-emitting devices 54 to 59 are different from those of the object detection system 100 of the first embodiment, and a part of the processing in the control device 20 is also different. Therefore, in the third embodiment, a description of the same configuration and the same control as those of the object detection system 100 of the first embodiment will be omitted, and a different configuration and control will be described.

The light emitting device 54 has five LEDs with different emission wavelengths. That is, the light emitting device 54 emits light with directivity. FIG. 6 shows a schematic diagram of the light emitting device 54 viewed from above. As shown in FIG. 6, the light emitting device 54 is placed on a curbstone ES (illustrated by a dashed line) located at the edge of the road. Further, the light emitting device 54 emits a green wavelength Gr in a direction perpendicular to the traveling direction of the vehicle 10 traveling on the road. The light emitting device 54 emits yellow wavelength Yr in directions forming angles α2 and α3 in both directions from the emission direction of wavelength Gr. Further, the light emitting device 54 emits a red wavelength Re in directions forming respective angles (α2+α1) and (α3+α4) in both directions from the light emission direction of the wavelength Gr. Each of the angles α1 to α4 in the third embodiment is 45 degrees.

FIG. 7 is a schematic diagram of a captured image IM2 including the light emitting devices 54-59 placed on the curb ES of the road RDa. As shown in FIG. 7, a plurality of light-emitting devices 54-59 are arranged at the edge of the curved road RDb. Each of the light emitting devices 55-59 is the same device as the light emitting device . In FIG. 7, for the sake of convenience, the direction of emission of green wavelength Gr from each of the light emitting devices 55 to 59 is indicated by a dashed line.

The target object DB 42 of the third embodiment stores the position of the curbstone corresponding to each of the light emitting devices 54 to 59 pre-associated with the light emission pattern of each of the light emitting devices 54 to 59 . As shown in FIG. 7, the emission wavelengths of the light emitting devices 55 to 59 captured by the onboard camera 11 differ depending on the positional relationship between the onboard camera 11 and the light emitting devices 55 to 59 . For example, since the light-emitting device 54 is positioned substantially on the side of the vehicle 10 , the camera 11 captures the green wavelength Gr as the light emission wavelength of the light-emitting device 54 . On the other hand, the direction of the green wavelength Gr in the light emitting device 58 forms an angle of about 45 degrees with the straight line connecting the light emitting device 58 and the vehicle 10 . Therefore, the camera 11 captures the yellow wavelength Yr as the emission wavelength of the light emitting device 58 .

The detection unit 33b of the third embodiment detects the wavelength (emission color) of each of the light emitting devices 55 to 59 captured by the in-vehicle camera 11 and the position of the curb corresponding to each of the light emitting devices 55 to 59 stored in the object DB 42b. and to detect the shape of the road RDb. As described above, in the third embodiment, the detection unit 33b uses the light emission patterns of the light emitting devices 54 to 59 to determine the positions of the light emitting devices 54 to 59 on the road RDb, as well as the directional light emission. , the orientation of each of the light emitting devices 55 to 59 is also detected. Therefore, the detection unit 33b of the third embodiment can more clearly detect the shape of the road RDb. As a result, the driver of the vehicle 10 can recognize the shape of the road on which the vehicle 10 is scheduled to travel, and can travel the road RDb more safely.

<Modified example of the third embodiment>
Various modifications can be made to the light emitting device 54 that emits light with directivity. For example, each angle α1-α4 shown in FIG. 6 may be a different angle. Further, it is not necessary to select wavelengths of colors that are symmetrical with respect to the direction of the green wavelength Gr, and the lengths of the wavelengths emitted in the five directions may all be different. Further, each of the light emitting devices 54 to 59 may emit light with different directivity, and may be set according to the degree of curve of the road RDb on which each of the light emitting devices 54 to 59 is installed. Further, each of the light emitting devices 54 to 59 may be installed even when the road RDb is straight.

<Fourth Embodiment>
FIG. 8 is an explanatory diagram showing a captured image IM4 of the vehicle-mounted camera 11 in the fourth embodiment. In the fourth embodiment, as compared with the first embodiment, the detection unit 33 detects the light emitting device 51c placed on the curb ES instead of the light emitting device 51 placed on the stop line HL. different. Therefore, in the fourth embodiment, a description of the same configuration and the same control as those of the first embodiment will be omitted, and a different configuration and control will be described.

The object DB 42 stores the position of the curbstone ES pre-associated with the light emission pattern of the light emitting device 51c. The position of the curbstone ES here refers to the position where the light-emitting device 51c is arranged among the curbstones ES that are linearly formed along the road RDc. The detection unit 33 detects that the light emitting device 51c represents the position of the curb by analyzing the temporal change in the light emission pattern of the light emitting device 51c in the captured image IM4. Therefore, the detection unit 33 of the fourth embodiment can clearly detect the curbstone ES that separates the road from the portion other than the road.

<Modification of Fourth Embodiment>
In the fourth embodiment, the light emitting device 51c is arranged at one of the curbstones ES formed linearly along the road RDc. can be modified in various ways. For example, as in the third embodiment, a plurality of light-emitting devices may be continuously arranged on the curbstone ES at predetermined intervals. By arranging a plurality of light-emitting devices on the curbstone ES in this way, the detection unit 33 can more clearly detect the curbstone ES that is continuously formed along the road RDc.

The present aspect has been described above based on the embodiments and modifications, but the above-described embodiments are intended to facilitate understanding of the present aspect, and do not limit the present aspect. This aspect may be modified and modified without departing from the spirit and scope of the claims, and this aspect includes equivalents thereof. Also, if the technical features are not described as essential in this specification, they can be deleted as appropriate.

DESCRIPTION OF SYMBOLS 10,10a... Vehicle 11... Vehicle-mounted camera (imaging part)
20, 20a... Control device 21... ROM
22 RAM
30 CPU
31... communication section 32... operation control section 33, 33a, 33b... detection section (target detection section)
34... Image processing unit (target detection unit)
40... Storage unit 42, 42a, 42b... Object DB
41 Map DB
51 to 59, 51c... Light emitting device 100, 100a... Object detection system B... Number of pixels W... Distance between light emitting devices Z... Distance from vehicle to crosswalk f... Focal length CL... Center line CW... Crosswalk ES... Curbstone HL...Stop line IM1, IM2, IM4...Captured image RD, RDb, RDc...Road Gr...Green wavelength Re...Red wavelength Yr...Yellow wavelength

Claims (7)

An object detection system comprising:
A light-emitting device arranged on or near an object provided on a road, the light-emitting device being capable of temporally modulating light-emitting conditions;
an imaging unit that is mounted on a vehicle and captures an image of the light emitting device positioned around the vehicle;
an object detection unit that detects an object pre-associated with the light emission condition of the light emitting device by analyzing changes over time in the captured image of the image pickup unit;
with
The light-emitting device is placed on the curb of the road and emits light with a wavelength having different directivity depending on the light-emitting direction,
The target detection unit is
detecting the position of the curb as an object;
detecting the shape of a road by using the position of each curb stone corresponding to each light emitting device and the directivity of each light emitting device when the captured image includes a plurality of the light emitting devices; detection system. An object detection system comprising:
A light-emitting device arranged on or near an object provided on a road, the light-emitting device being capable of temporally modulating light-emitting conditions;
an imaging unit that is mounted on a vehicle and captures an image of the light emitting device positioned around the vehicle;
an object detection unit that detects an object pre-associated with the light emission condition of the light emitting device by analyzing changes over time in the captured image of the image pickup unit;
with
The light-emitting devices are arranged at least two locations on a road crosswalk,
The target detection unit is
detecting the position of the pedestrian crossing as an object;
An object detection system that calculates a distance from the vehicle to the crosswalk using the distances between the light emitting devices in the captured image. The object detection system according to claim 1 or claim 2 ,
The light emitting device is arranged on a road stop line,
The object detection system, wherein the object detection unit detects at least one of the stop line as the object and a position of a pedestrian crossing arranged near the stop line as the object. The object detection system according to any one of claims 1 to 3 ,
The light emission condition is a pulsed light emission pattern in which light emission by the light emitting device is repeatedly turned on and off over time,
The object detection system, wherein the emission pattern varies at least one of a pulse wavelength, a pulse emission intensity, and a pulse width and period. The object detection system according to claim 4 ,
The light emission pattern of the light emitting device is set according to the type of object with which the light emitting device is associated,
The object detection system, wherein the object detection unit distinguishes and detects different objects by analyzing synchronization of the light emission patterns of the light emitting devices in the captured image. An object detection method comprising:
A light-emitting device that is placed on or near an object placed on a road and whose light-emitting conditions can be modulated over time, and that is placed on a curb of the road and emits wavelengths that have different directivities depending on the light-emitting direction. an image capturing step of capturing an image of the light emitting device that emits light using an image capturing unit mounted on the vehicle;
an analysis step of analyzing changes over time of the captured image of the imaging unit;
a detection step of detecting an object pre-associated with the light emission condition of the light emitting device using the analysis result ;
The detection step includes
detecting the position of the curb as an object;
detecting the shape of a road using the position of each curbstone corresponding to each light emitting device and the directivity of each light emitting device when the captured image includes a plurality of the light emitting devices; Detection method. An object detection method comprising:
A light-emitting device that is arranged on or near an object provided on a road and whose light-emitting conditions can be modulated over time, and that is arranged in at least two places on a crosswalk on the road is mounted on a vehicle. an imaging step of imaging with an imaging unit;
an analysis step of analyzing changes over time of the captured image of the imaging unit;
a detection step of detecting an object pre-associated with the light emission condition of the light emitting device using the analysis result;
The detection step includes
detecting the position of the pedestrian crossing as an object;
An object detection method, wherein the distance from the vehicle to the pedestrian crossing is calculated using the distances between the light emitting devices in the captured image.

Images