JP7245653B2 - Traffic light recognition method and traffic light recognition device - Google Patents

Description

The present invention relates to a traffic signal recognition method and a traffic signal recognition device.

As a vehicle exterior recognition device that recognizes a traffic signal from a front camera image that captures the front of the vehicle, the vehicle's lane is detected based on the detection target outside the vehicle (traffic light, stop sign, etc.) detected from the front camera image and the road structure. A vehicle exterior recognition device for recognizing a traffic signal has been proposed (see Patent Document 1).

JP 2007-257299 A

However, according to the technique described in Patent Document 1, when a traffic signal to be recognized is blocked by the presence of an object around the own vehicle, there is a risk that the traffic signal cannot be recognized based on the front camera image. , there may be a case where the state of the traffic light cannot be continuously recognized in the section from the far side of the intersection to the intersection.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and its object is to reduce the influence of shielding of traffic lights by objects around the own vehicle, and to reduce the effects of traffic light shielding by objects around the own vehicle when the own vehicle travels from a distance to an intersection. To provide a traffic light recognition method and a traffic light recognition device capable of continuously recognizing the state of the traffic light.

In order to solve the above-described problems, a traffic light recognition method and a traffic light recognition device according to an aspect of the present invention are based on the position of a vehicle, the position of a traffic signal in front of the vehicle, and the presence or absence of an object in front of the vehicle. , while the vehicle is traveling to the position of the stop line, it is determined whether or not the traffic signal can be blocked, and when it is determined that the traffic signal can be blocked, the front of the vehicle is imaged by the first imaging unit The first image is selected as a processed image from among the first image and a second image captured by a second imaging unit having a narrower angle of view than the first image at a distance farther than the imageable range of the first image, and is used as the processed image. based on the detection of traffic lights.

Advantageous Effects of Invention According to the present invention, it is possible to continuously recognize the state of a traffic signal when the vehicle travels from a distance to an intersection while reducing the influence of the obstruction of the traffic signal by objects around the vehicle.

FIG. 1 is a block diagram showing the configuration of a traffic light recognition device according to one embodiment of the present invention. FIG. 2 is a flow chart showing the processing procedure of the traffic light recognition device according to one embodiment of the present invention. FIG. 3 is a schematic diagram showing switching positions of processed images when there is no preceding vehicle or adjacent vehicle. FIG. 4A is a schematic diagram showing switching positions of processed images when there is a preceding vehicle. FIG. 4B is a schematic diagram showing the switching position of the processed image when there is an adjacent vehicle.

Embodiments of the present invention will now be described in detail with reference to the drawings. In the explanation, the same reference numerals are given to the same parts, and redundant explanations are omitted.

[Configuration of Traffic Light Recognition Device]
FIG. 1 is a block diagram showing the configuration of a traffic light recognition device according to this embodiment. As shown in FIG. 1, the traffic light recognition device 1 according to the present embodiment includes a wide-angle camera 22 (first imaging unit), a telephoto camera 24 (second imaging unit), and a controller 100. The controller 100 Controller 100 is connected to vehicle control device 200 . The controller 100 also includes a traffic light position acquisition unit 52, a surrounding object position calculation unit 54, an own vehicle position acquisition unit 56, a blocking possibility determination unit 60, a processed image selection unit 70, a traffic light detection unit 80, and an output unit 90 . The traffic light recognition device 1 is mounted on a vehicle (not shown).

The wide-angle camera 22 and the telephoto camera 24 are digital cameras equipped with solid-state imaging devices such as CCD and CMOS, and acquire digital images of the front of the vehicle by imaging the front of the vehicle. The wide-angle camera 22 and the telephoto camera 24 capture a predetermined range around the vehicle by setting the focal length, the angle of view of the lens, the vertical and horizontal camera angles, and the like. The wide-angle camera 22 and the telephoto camera 24 output the acquired captured images to the processing image selection unit 70, which will be described later.

Compared to the wide-angle camera 22, the telephoto camera 24 captures a narrower range of angle of view than the wide-angle camera 22, but can capture an area farther than the wide-angle camera 22 can capture. More specifically, the wide-angle camera 22 is set to have a short focal length so as to be able to capture an image of the vicinity of the vehicle, and is capable of capturing a wide-angle image. The telephoto camera 24 is set to have a long focal length so that it can capture an image farther than the range that can be captured by the wide-angle camera 22, and can capture an image at a narrow angle.

The wide-angle camera 22 and the telephoto camera 24 are desirably mounted on the front portion of the vehicle so that images of the front of the vehicle can be captured.

In addition, traffic signals in a specific direction cannot be seen from the wide-angle camera 22 and the telephoto camera 24 at the same time due to a preceding vehicle traveling in front of the own vehicle or an adjacent vehicle traveling on a lane adjacent to the lane in which the own vehicle is traveling. In order to prevent (occlusion), the wide-angle camera 22 and the telephoto camera 24 may be arranged as far apart as possible. For example, the wide-angle camera 22 is installed at a position close to the eye position of the driver of the vehicle, and the telephoto camera 24 is installed at a position on the passenger seat side, away from the installation position of the wide-angle camera 22 in the width direction of the vehicle. It may be arranged.

Note that when a plurality of traffic lights can be recognized by the wide-angle camera 22 (for example, when a traffic light installed above the lane on which the vehicle is traveling can be recognized, as well as a traffic light installed above the oncoming lane), If so, the problem of occlusion is less likely to occur. In this case, the wide-angle camera 22 and the telephoto camera 24 may be arranged at the same place or arranged close to each other.

The captured images captured by the wide-angle camera 22 and the telephoto camera 24 may be stored in a storage unit (not shown) for a predetermined period of time. For example, the wide-angle camera 22 and telephoto camera 24 acquire captured images at predetermined time intervals, and the captured images acquired at predetermined time intervals are stored in the storage unit as past images. The past image may be deleted after a predetermined period has passed since the time when the past image was captured.

The traffic light position acquiring unit 52 acquires the position of the traffic light within the road structure around the own vehicle. The traffic light position acquisition unit 52 may acquire the position of the traffic light in the road structure based on the position and posture of the vehicle and map information, or may perform image analysis on the captured image acquired in the past. By doing so, the position of the traffic light within the road structure may be obtained.

The map information may be obtained by a map obtaining unit (not shown). The map acquisition unit may own a map database storing map information, or may acquire map information from an external map data server by cloud computing. The map information acquired by the map acquisition unit includes road structure information such as the absolute position of lanes, the connection relationship between lanes, the relative positional relationship, the position of a stop line, the position of a traffic signal corresponding to a stop line, and the type of traffic signal. be

The surrounding object position calculator 54 includes an object detection sensor such as a laser radar, a millimeter wave radar, and a camera mounted on the own vehicle. The surrounding object position calculator 54 detects objects around the vehicle using a plurality of object detection sensors. In addition, the surrounding object position calculator 54 detects an object in front of or on the side of the own vehicle. The surrounding object position calculator 54 detects moving objects including other vehicles, motorcycles, bicycles and pedestrians, and stationary objects including parked vehicles and buildings. For example, the surrounding object position calculator 54 detects the position, attitude (yaw angle), size, speed, acceleration, jerk, deceleration, and yaw rate of a moving object and a stationary object relative to the own vehicle.

The own vehicle position acquisition unit 56 includes a position detection sensor such as GPS (Global Positioning System) and odometry mounted on the own vehicle that measures the absolute position of the own vehicle. The own vehicle position acquisition unit 56 uses a position detection sensor to measure the absolute position of the own vehicle, that is, the position, vehicle speed, acceleration, steering angle, and attitude of the own vehicle with respect to a predetermined reference point. The vehicle position acquisition unit 56 includes an inertial navigation system (INS), sensors provided on the brake pedal and accelerator pedal, sensors for acquiring vehicle behavior such as wheel-side sensors and yaw rate sensors, and a laser. Radar, cameras, etc. may be included.

Based on the position of the traffic light and the presence or absence of objects around the vehicle, the shielding possibility determination unit 60 determines that, while the vehicle travels along the travel route from the position of the vehicle to the position of the stop line corresponding to the traffic light, It is determined whether or not the traffic signal can be blocked by the object detected by the surrounding object position calculator 54 . Specifically, the shielding possibility determination unit 60 acquires the position of the stop line corresponding to the traffic light, which exists on the road structure around the own vehicle. While the vehicle travels along the route connecting the current position of the vehicle and the position of the acquired stop line, the traffic lights seen from the wide-angle camera 22 and the telephoto camera 24 mounted on the vehicle are shielded. determine whether or not

Here, similar to the acquisition of the position of the traffic signal by the traffic signal position acquisition unit 52, the position of the stop line corresponding to the traffic signal may be acquired based on the position and attitude of the own vehicle and the map information. Alternatively, the position of the stop line corresponding to the traffic light may be acquired by image analysis of the captured image acquired in the past.

Further, when the shielding possibility determination unit 60 determines whether or not the traffic light is shielded, in addition to the information on the presence or absence of objects around the own vehicle, the positions and sizes of the objects around the own vehicle are used. , width and height information may be used. In addition, the shielding possibility determination unit 60 calculates the expected behavior of the object, such as the assumed motion trajectory of the object around the vehicle, and determines whether the traffic light will be shielded based on the expected behavior. It may be determined whether

The processing image selection unit 70 selects the wide-angle camera 22 based on the distance between the stop line and the own vehicle, the determination result of the shielding possibility determination unit 60, and the imaging range of the wide-angle camera 22 and the telephoto camera 24. , or a telephoto image (second image) captured by the telephoto camera 24 is selected as a "processed image" to be processed for traffic signal detection. A method of selecting the "processed image" will be described later.

The traffic light detector 80 detects a traffic light based on the processed image selected by the processed image selector 70 . The traffic light detection unit 80 sets a detection area on the processed image for each traffic signal that is expected to appear in the processed image, and analyzes the pixels in the set detection area. It may be one that detects a traffic light that is

The output unit 90 outputs the result of the traffic signal detection by the traffic signal detection unit 80 to the vehicle control device 200 .

[Method of selecting images to be processed]
Next, a method of selecting a "processed image" performed by the processed image selection unit 70 will be described with reference to FIGS. 3, 4A, and 4B.

First, a method of selecting a processed image when it is considered that there is no possibility that the traffic light will be blocked by preceding or adjacent vehicles will be described. FIG. 3 is a schematic diagram showing the switching position PC of the processed image when there is no preceding vehicle or adjacent vehicle. The position P0 is the position of the stop line corresponding to the traffic signal TS, and the position PA is the position where the signal recognition process is started (for example, the position where the signal recognition process based on the telephoto image is possible and is the farthest from the traffic signal TS).

Here, the position PC is determined by the position of the traffic signal TS and the angle of view of the telephoto camera 24 on the travel route of the vehicle V1 from the position PA to the position P0, and is determined from the positions where the telephoto camera 24 can capture the traffic signal TS. , the position closest to the position P0. When the own vehicle V1 advances to a position closer to the position P0 than the position PC, the traffic light TS is out of the angle of view of the telephoto camera 24. FIG. The traffic light TS cannot be imaged by the telephoto camera 24.

If there is no preceding vehicle or adjacent vehicle and there is no possibility that the traffic light TS will be blocked as seen from the telephoto camera 24 while the host vehicle V1 travels from the position PA to the position PC, the processed image selection unit 70 The telescopic image is selected as the processed image while the vehicle V1 is traveling from the position PA to the position PC, and the wide-angle image is selected as the processed image while the vehicle V1 is traveling from the position PC to the position P0. That is, the processed image selection unit 70 switches the processed image from the telescopic image to the wide-angle image at the position PC.

In other words, the processed image selection unit 70 sets the distance from the position PC to the position P0 as the second distance, and if the distance between the position P0 where the stop line is located and the host vehicle V1 is less than or equal to the second distance, , select the wide-angle image as the processed image.

The reason for this is that the traffic light TS is out of the angle of view of the telephoto camera 24 while the host vehicle V1 is traveling from the position PC to the position P0. This is because the traffic light TS cannot be imaged by the telephoto camera 24 . Therefore, the wide-angle image is selected as the image to be processed so that the detection of the traffic light is possible.

On the other hand, if the distance between the stop line position P0 and the own vehicle V1 is greater than the second distance, the processing image selection unit 70 selects the telescopic image as the processing image. That is, when there is no possibility that the traffic signal TS will be blocked as seen from the telephoto camera 24 while the host vehicle V1 is traveling from the position PA to the position PC, the processed image selection unit 70 reaches a position as close to the position PC as possible. Until then, the telephoto image is selected as the processed image so that traffic light detection can be performed based on the telephoto image.

Next, a method of selecting a processed image when there is a possibility that a traffic signal is blocked by a preceding vehicle will be described. FIG. 4A is a schematic diagram showing the switching position PS of the processed image when the preceding vehicle V2 is present.

Here, the position PB is determined by the position of the traffic signal TS and the angle of view of the wide-angle camera 22 on the travel route of the vehicle V1 from the position PA to the position P0, and is determined from the positions where the traffic signal TS can be imaged by the wide-angle camera 22. , is the farthest position from the position P0. When the own vehicle V1 is located at a position farther from the position P0 than the position PB, the distance to the traffic light TS is large, so that the traffic light TS has a pixel count equal to or less than a predetermined pixel count in the wide-angle image captured by the wide-angle camera 22. As a result, the traffic light TS cannot be recognized from the wide-angle image.

If there is a preceding vehicle V2 and there is a possibility that the traffic signal TS will be blocked as seen from the vehicle V1 (as seen from the telephoto camera 24) while the vehicle V1 travels from the position PA to the position PC, the processed image The selector 70 determines the position PS located in the section between the position PC and the position PB (the section including the position PC and the position PB). The telephoto image is selected as the processed image while the vehicle V1 is traveling from the position PA to the position PS, and the wide-angle image is selected as the processed image while the vehicle V1 is traveling from the position PS to the position P0. That is, the processing image selection unit 70 switches the processing image from the telephoto image to the wide-angle image at the position PS.

It should be noted that if the traffic signal TS is not shielded, it is possible to capture the traffic signal TS in both the telephoto image and the wide-angle image in the section between the positions PC and PB.

In other words, the processed image selection unit 70 sets the distance from the position PB to the position P0 as the first distance, and if the distance between the position P0 with the stop line and the host vehicle V1 is greater than the first distance, , the processing image selection unit 70 selects a telescopic image as a processing image.

On the other hand, when the distance between the position P0 where the stop line is located and the host vehicle V1 is equal to or less than the second distance, which is the distance from the position PC to the position P0, the processing image selection unit 70 selects the telephoto image as the processing image. Select as

Further, when the distance between the position P0 of the stop line and the own vehicle V1 is greater than the second distance and equal to or less than the first distance, the processed image is changed from the telescopic image to the wide-angle image at the predetermined position PS. switch. This is because when there is a possibility that the traffic signal TS will be blocked by the preceding vehicle V2, using the wide-angle image as the processed image is less likely to be affected by the blocking than the telephoto image, and the possibility that the traffic signal can be continuously detected is high. Because it becomes

When the own vehicle V1 does not automatically follow the preceding vehicle V2, it is necessary to continue detecting the traffic signal TS as much as possible while suppressing the influence of the preceding vehicle V2 shielding the traffic signal TS. Therefore, it is desirable that the processing image selection unit 70 switches the processing image from the telephoto image to the wide-angle image at the timing when the own vehicle V1 is positioned at the position PB.

On the other hand, when the own vehicle V1 is automatically following the preceding vehicle V2, even if the traffic signal TS cannot be detected, the own vehicle V1 can start and stop based on the traveling of the preceding vehicle V2. There is little need to switch the processed image from the telephoto image to the wide-angle image at the timing when the own vehicle V1 is positioned at PB. Therefore, the processing image selection unit 70 may switch the processing image from the telescopic image to the wide-angle image while the vehicle is traveling toward the position PC after passing the position PB.

Next, a method of selecting a processed image when it is considered that there is a possibility that the traffic signal is blocked by an adjacent vehicle will be described. FIG. 4B is a schematic diagram showing the switching position PB of the processed image when there is an adjacent vehicle V3.

If there is an adjacent vehicle V3 and there is a possibility that the traffic light TS will be blocked from the viewpoint of the own vehicle V1 while the own vehicle V1 is traveling from the position PA to the position P0, the processed image selection unit 70 selects the signal TS of the own vehicle V1 selects a telescopic image as a processed image while V1 travels from position PA to position PB, and selects a wide-angle image as a processed image while vehicle V1 travels from position PB to position P0. That is, the processed image selection unit 70 switches the processed image from the telephoto image to the wide-angle image at the position PB.

The reason for this is that, for example, when the road on which the vehicle V1 is traveling is curved, there is a possibility that the traffic signal will be blocked by the adjacent vehicle V3 traveling in the lane adjacent to the lane on which the vehicle V1 is traveling. , it is necessary to continuously detect the traffic signal TS as much as possible.

[Processing procedure of traffic light recognition device]
Next, a processing procedure for traffic signal recognition by the traffic signal recognition device 1 according to the present embodiment will be described with reference to the flowchart of FIG. The signal recognition processing shown in FIG. 2 is started when the ignition of the vehicle is turned on, and is repeatedly executed while the ignition is on.

In step S101, the telephoto camera 24 images a predetermined range around the vehicle. The telephoto camera 24 outputs the acquired telephoto image to the processed image selection unit 70 .

In step S103, the wide-angle camera 22 images a predetermined range around the vehicle. The wide-angle camera 22 outputs the acquired wide-angle image to the processed image selection section 70 .

The setting information (focal length, lens angle of view, camera vertical and horizontal angles, etc.) in the wide-angle camera 22 and the telephoto camera 24 may be set in advance, or may be set in accordance with a command from the controller 100. , may be set each time. The captured images captured by the wide-angle camera 22 and the telephoto camera 24 may be stored in a storage unit (not shown) for a predetermined period of time.

In step S105, the own vehicle position acquisition unit 56 acquires the position of the own vehicle.

In step S107, the traffic signal position acquiring unit 52 acquires the position of the traffic signal within the road structure around the vehicle.

In step S109, the surrounding object position calculator 54 detects objects around the host vehicle. In particular, the surrounding object position calculator 54 detects an object in front of the vehicle, and also detects the position, speed, etc. of the detected object.

In step S111, based on the position of the traffic light and the presence or absence of objects around the vehicle, the shielding possibility determination unit 60 determines whether the vehicle is traveling along the travel route from the position of the vehicle to the position of the stop line corresponding to the traffic signal. In the meantime, it is determined whether or not the traffic signal can be blocked by the object detected by the surrounding object position calculator 54 .

In step S113, the processed image selection unit 70 selects the image based on the distance between the stop line and the host vehicle, the determination result of the shielding possibility determination unit 60, and the imaging range of the wide-angle camera 22 and the telephoto camera 24. , one of the wide-angle image and the telephoto image is selected as a "processed image" to be subjected to traffic signal detection processing.

In step S<b>115 , the traffic signal detection unit 80 detects a traffic signal based on the processed image selected by the processed image selection unit 70 .

In step S<b>117 , the output unit 90 outputs the signal detection result by the signal detection unit 80 to the vehicle control device 200 .

[Effects of Embodiment]
As described in detail above, the traffic signal recognition method and the traffic signal recognition device according to the present embodiment include a first image captured by a first imaging unit that is mounted on the vehicle and captures an image of the front of the vehicle; is a second imaging unit that captures an image of the and/or an image is selected as a processed image, and when recognizing a traffic signal in front of the vehicle based on the processed image, the position of the vehicle and the position of the traffic signal are obtained, and an object in front of the vehicle is detected. Then, based on the position of the traffic light and the presence or absence of the object, while the own vehicle is traveling along the travel route from the position of the own vehicle to the position of the stop line corresponding to the traffic light, when the second image is selected, the object determines whether or not the traffic light can be blocked as viewed from the second imaging unit, and if it is determined that the traffic light can be blocked as viewed from the second imaging unit, the first image is selected as the processed image, and processed Detect traffic lights based on images.

Thereby, when there is a possibility that the traffic signal is blocked by an object in front of the own vehicle, the processed image can be changed from the second image to the first image, and the traffic signal can be continuously detected. As a result, it is possible to reduce the influence of traffic signal shielding due to the presence or absence of objects on the performance of traffic signal recognition.

In addition, the traffic signal recognition method and the traffic signal recognition device according to the present embodiment can detect a traffic signal at a position on a travel route where the traffic signal can be captured by the first imaging unit, which is determined by the position of the traffic signal and the angle of view of the first imaging unit. The distance to the stop line at the farthest position from the line is defined as a first distance, and if the distance between the stop line and the host vehicle is greater than the first distance, the second image is selected as the processed image. may

Thereby, the traffic signal can be detected based on the second image captured by the second imaging section capable of capturing a distance longer than the distance that can be captured by the first imaging section. That is, it is possible to switch between the two imaging units according to the distance to the stop line in consideration of the characteristics of the two imaging units having different angles of view. As a result, when the own vehicle travels toward the stop line, the traffic light can be continuously detected in a continuous section from a place far to the stop line to a place close to the stop line. In particular, when the traffic light cannot be recognized from the first image as a result of the traffic light having a number of pixels equal to or less than a predetermined number of pixels in the first image captured by the first imaging unit because the distance to the stop line is large, A second image can be used in place of the first image to perform traffic light recognition.

Furthermore, the traffic signal recognition method and the traffic signal recognition device according to the present embodiment are arranged so that the traffic signal can be imaged by the second imaging unit on the travel route, which is determined by the position of the traffic signal and the angle of view of the second imaging unit. The second distance is the distance to the stop line closest to the line, and the first image is selected as the processed image when the distance between the stop line and the host vehicle is less than or equal to the second distance. may

Accordingly, the traffic signal can be detected based on the first image captured by the first imaging section capable of capturing a distance shorter than the distance that can be captured by the second imaging section. That is, it is possible to switch between the two imaging units according to the distance to the stop line in consideration of the characteristics of the two imaging units having different angles of view. As a result, when the host vehicle travels toward the traffic signal, the traffic signal can be continuously detected in a continuous section from a place where the distance to the stop line is far to a place where the distance is close. In particular, when the distance to the stop line is too close to the angle of view of the second imaging unit, the traffic light can be recognized using the first image instead of the second image.

Further, the traffic signal recognition method and the traffic signal recognition device according to the present embodiment may detect, as an object, a preceding vehicle traveling in the same lane as the own vehicle is traveling. There is a risk that the preceding vehicle may block the traffic light that should appear in the image captured by the second imaging unit. By identifying the preceding vehicle as the object blocking the traffic light, the information necessary to determine whether the traffic light can be blocked can be accurately obtained.

Further, in the traffic signal recognition method and the traffic signal recognition device according to the present embodiment, when it is determined that the traffic signal can be blocked by the preceding vehicle, the distance between the stop line and the own vehicle is greater than the second distance, and The processed image may be switched from the second image to the first image while the distance is equal to or less than the first distance. When there is a possibility that the traffic signal is blocked by the preceding vehicle, the first image obtained by capturing an area close to the own vehicle at a wide angle is used as the processed image rather than the second image obtained by capturing an area far from the own vehicle at a narrow angle. The wide angle makes it less susceptible to traffic signal shielding by preceding vehicles. Since it is not easily affected by shielding, it is possible to increase the possibility of continuously detecting a traffic light.

Further, in the traffic signal recognition method and the traffic signal recognition device according to the present embodiment, when it is determined that the traffic signal can be blocked by the preceding vehicle and the host vehicle is not automatically following the preceding vehicle, The processed image may be switched from the second image to the first image at the timing when the distance between the stop line and the own vehicle becomes the first distance. By starting recognition of the traffic light using the first image from the farthest position from the stop line where the traffic light can be recognized from the first image when the own vehicle travels toward the stop line. , the traffic signal can be continuously detected while suppressing the influence of the shielding of the traffic signal by the preceding vehicle.

Furthermore, the traffic signal recognition method and the traffic signal recognition device according to the present embodiment may detect, as an object, an adjacent vehicle traveling in a lane adjacent to the lane in which the host vehicle is traveling. An adjacent vehicle may block the traffic signal that should appear in the images captured by the first imaging unit and the second imaging unit. By identifying adjacent vehicles as objects blocking the traffic light, the information necessary to determine whether the traffic light can be blocked can be accurately obtained.

Further, in the traffic signal recognition method and the traffic signal recognition device according to the present embodiment, when it is determined that the traffic signal may be blocked by an adjacent vehicle, the processed image may be switched from the second image to the first image. By starting recognition of the traffic light using the first image from the farthest position from the stop line where the traffic light can be recognized from the first image when the own vehicle travels toward the stop line. , the traffic signal can be continuously detected while suppressing the influence of the shielding of the traffic signal by adjacent vehicles. When the traffic signal can be blocked by the adjacent vehicle, the traffic signal can be recognized in the lane opposite to the lane in which the adjacent vehicle is traveling across the lane in which the own vehicle is traveling, and the traffic signal can be continuously recognized. be able to.

Furthermore, in the traffic light recognition method and the traffic light recognition device according to the present embodiment, the focal length of the first imaging section may be shorter than the focal length of the second imaging section. The area imaged by the first imaging unit and the area imaged by the second imaging unit are different, and by recognizing the traffic signal by switching between the first imaging unit and the second imaging unit, the distance to the stop line corresponding to the traffic signal can be increased. A traffic signal can be continuously detected in a continuous section from a far place to a near place.

Further, in the traffic signal recognition method and the traffic signal recognition device according to the present embodiment, the first imaging unit is installed at a position closer to the eyes of the driver of the own vehicle than the second imaging unit. There may be. As a result, the image captured by the first imaging unit, which can capture a wide-angle area compared to the second imaging unit, is similar to the image seen from the viewpoint of the driver of the own vehicle, and the traffic signal is detected based on the first image. is recognized, and when driving control and driving assistance are performed, the sense of discomfort given to the driver can be reduced.

Each function illustrated in the above embodiments may be implemented by one or more processing circuits. Processing circuitry includes programmed processors, electrical circuits, etc., as well as devices such as application specific integrated circuits (ASICs) and circuit components arranged to perform the described functions. etc. are also included.

Although the contents of the present invention have been described above according to the embodiments, it is obvious to those skilled in the art that the present invention is not limited to these descriptions and that various modifications and improvements are possible. The discussion and drawings forming part of this disclosure should not be understood as limiting the invention. Various alternative embodiments, implementations and operational techniques will become apparent to those skilled in the art from this disclosure.

Of course, the present invention includes various embodiments and the like that are not described here. Therefore, the technical scope of the present invention is defined only by the matters specifying the invention according to the valid scope of claims based on the above description.

1 Traffic light recognition device 1
22 wide-angle camera (first imaging unit)
24 telephoto camera (second imaging unit)
52 Traffic light position acquisition unit 54 Surrounding object position calculation unit 56 Own vehicle position acquisition unit 60 Shielding possibility determination unit 70 Processed image selection unit 80 Traffic light detection unit 90 Output unit 100 Controller 100
V1 Own vehicle V2 Leading vehicle V3 Adjacent vehicle

Claims (8)

a first imaging unit that is mounted on the vehicle and captures an image in front of the vehicle;
A second image pickup unit mounted on a vehicle for picking up an image of the front of the vehicle, the second image pickup unit having an angle of view narrower than that of the first image pickup unit and picking up an image farther than the imageable range of the first image pickup unit when viewed from the own vehicle. the second imaging unit;
a controller;
A traffic light recognition method according to a traffic light recognition device comprising
The controller is
selecting one of the first image captured by the first imaging unit and the second image captured by the second imaging unit as a processed image, and detecting a traffic light based on the processed image;
Acquiring the position of the own vehicle and the position of the traffic light in front of the own vehicle;
Detecting a preceding vehicle traveling in the same lane ahead of the own vehicle as the lane in which the own vehicle is traveling;
Based on the position of the vehicle, the position of the traffic signal, and the presence or absence of the preceding vehicle, it is determined whether or not the traffic signal can be blocked while the vehicle travels to the position of the stop line corresponding to the traffic signal. death,
The first imaging unit can image the traffic light on the travel route from the position of the vehicle to the position of the stop line corresponding to the traffic light, which is determined by the position of the traffic light and the angle of view of the first imaging unit. position, the distance to the stop line at the farthest position from the stop line as a first distance,
When it is determined that the traffic light can be blocked by the preceding vehicle when the second image is selected, and when the host vehicle is not automatically following the preceding vehicle, Switching the processed image from the second image to the first image at the timing when the distance between the stop line and the host vehicle becomes the first distance.
A traffic light recognition method characterized by: a first imaging unit that is mounted on the vehicle and captures an image in front of the vehicle;
A second image pickup unit mounted on a vehicle for picking up an image of the front of the vehicle, the second image pickup unit having an angle of view narrower than that of the first image pickup unit and picking up an image farther than the imageable range of the first image pickup unit when viewed from the own vehicle. the second imaging unit;
a controller;
A traffic light recognition method according to a traffic light recognition device comprising
The controller is
selecting one of the first image captured by the first imaging unit and the second image captured by the second imaging unit as a processed image, and detecting a traffic light based on the processed image;
Acquiring the position of the own vehicle and the position of the traffic light in front of the own vehicle;
Detecting an adjacent vehicle traveling in a lane in front of the own vehicle and adjacent to the lane in which the own vehicle travels;
Based on the position of the vehicle, the position of the traffic signal, and the presence or absence of the adjacent vehicle, it is determined whether or not the traffic signal can be blocked while the vehicle travels to the position of the stop line corresponding to the traffic signal. death,
The first imaging unit can image the traffic light on the travel route from the position of the vehicle to the position of the stop line corresponding to the traffic light, which is determined by the position of the traffic light and the angle of view of the first imaging unit. position, the distance to the stop line at the farthest position from the stop line as a first distance,
When it is determined that the traffic light may be blocked by the adjacent vehicle while the second image is being selected, at the timing when the distance between the stop line and the own vehicle becomes the first distance, the switching the processed image from the second image to the first image;
A traffic light recognition method characterized by: 3. The traffic light recognition method according to claim 1 , wherein the controller processes the second image when the distance between the stop line and the host vehicle is greater than the first distance. A traffic light recognition method characterized by selecting as an image. The traffic light recognition method according to claim 3 ,
The controller is
To the stop line closest to the stop line at a position on the travel route where the traffic light can be imaged by the second imaging unit, which is determined by the position of the traffic light and the angle of view of the second imaging unit. as the second distance,
A traffic light recognition method, wherein the first image is selected as the processed image when the distance between the stop line and the host vehicle is equal to or less than the second distance. The traffic light recognition method according to any one of claims 1 to 4 ,
The traffic light recognition method, wherein the focal length of the first imaging section is shorter than the focal length of the second imaging section. The traffic light recognition method according to any one of claims 1 to 5 ,
The traffic signal recognition method, wherein the first imaging unit is installed at a position closer to the eyes of the driver of the vehicle than the second imaging unit. a first imaging unit that is mounted on the vehicle and captures an image in front of the vehicle;
A second imaging unit mounted on a vehicle for capturing an image of the front of the vehicle, the second imaging unit having a narrower angle of view than the first imaging unit and capturing an image farther from the vehicle than the imageable range of the first imaging unit. the second imaging unit;
a controller;
A traffic light recognition device comprising:
The controller is
selecting one of the first image captured by the first imaging unit and the second image captured by the second imaging unit as a processed image, and detecting a traffic light based on the processed image;
Acquiring the position of the own vehicle and the position of the traffic light in front of the own vehicle;
Detecting a preceding vehicle traveling in the same lane ahead of the own vehicle as the lane in which the own vehicle is traveling;
Based on the position of the vehicle, the position of the traffic signal, and the presence or absence of the preceding vehicle, it is determined whether or not the traffic signal can be blocked while the vehicle travels to the position of the stop line corresponding to the traffic signal. death,
The first imaging unit can image the traffic light on the travel route from the position of the vehicle to the position of the stop line corresponding to the traffic light, which is determined by the position of the traffic light and the angle of view of the first imaging unit. position, the distance to the stop line at the farthest position from the stop line as a first distance,
When it is determined that the traffic light can be blocked by the preceding vehicle when the second image is selected, and when the host vehicle is not automatically following the preceding vehicle, Switching the processed image from the second image to the first image at the timing when the distance between the stop line and the host vehicle becomes the first distance.
A traffic light recognition device characterized by: a first imaging unit that is mounted on the vehicle and captures an image in front of the vehicle;
A second imaging unit mounted on a vehicle for capturing an image of the front of the vehicle, the second imaging unit having a narrower angle of view than the first imaging unit and capturing an image farther from the vehicle than the imageable range of the first imaging unit. the second imaging unit;
a controller;
A traffic light recognition device comprising:
The controller is
selecting one of the first image captured by the first imaging unit and the second image captured by the second imaging unit as a processed image, and detecting a traffic light based on the processed image;
Acquiring the position of the own vehicle and the position of the traffic light in front of the own vehicle;
Detecting an adjacent vehicle traveling in a lane in front of the own vehicle and adjacent to the lane in which the own vehicle travels;
Based on the position of the vehicle, the position of the traffic signal, and the presence or absence of the adjacent vehicle, it is determined whether or not the traffic signal can be blocked while the vehicle travels to the position of the stop line corresponding to the traffic signal. death,
The first imaging unit can image the traffic light on the travel route from the position of the vehicle to the position of the stop line corresponding to the traffic light, which is determined by the position of the traffic light and the angle of view of the first imaging unit. position, the distance to the stop line at the farthest position from the stop line as a first distance,
When it is determined that the traffic light may be blocked by the adjacent vehicle while the second image is being selected, at the timing when the distance between the stop line and the own vehicle becomes the first distance, the switching the processed image from the second image to the first image;
A traffic light recognition device characterized by:

Images