JP2012073927A - Driving support apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driving support apparatus capable of preventing delay of recognition, when the driving support apparatus recognizes traveling environment of an own vehicle by processing a camera image, regardless of a sudden change of a photographing condition of a camera from a bright environment to a dark environment.SOLUTION: A recognition controller of a driving support apparatus comprises a tunnel recognition part 22a and an exposure adjustment part 22b. When the tunnel recognition part 22a recognizes a tunnel in front of an own vehicle through a captured image by a stereoscopic camera, it stores the distance from the own vehicle to the tunnel entrance. After the recognition result is transmitted from the tunnel recognition part 22a to the exposure adjustment part 22b, the exposure adjustment part 22b switches shutter speed of the stereoscopic camera at the timing determined based on the distance from the own vehicle to the tunnel entrance, adjusts the exposure amount of the captured image of the stereoscopic camera taken right after entering the tunnel, and avoids delays of the recognition.

Description

  The present invention relates to a driving support apparatus that performs stereo processing on an image captured by a stereo camera mounted on a vehicle, recognizes a driving environment, and performs driving support for a driver.

  In recent years, as a vehicle driving support device, an on-vehicle stereo camera or a monocular camera captures a forward driving environment, performs image recognition processing, and performs alarm control for a front obstacle. A technology for detecting the vehicle and performing follow-up control and warning control on the preceding vehicle has been developed and put into practical use.

  In such driving assistance using a camera image, the driving environment of the vehicle changes in various ways. If the dynamic range of the camera is insufficient, saturation of the bright part of the image (blackout) and blackening of the dark part occur. The image recognition may be difficult.

  For this reason, for example, Patent Document 1 captures images while periodically changing the shutter speed or lens aperture of a camera set in advance, and the subject is appropriately captured from the plurality of captured images. A technique is disclosed in which a subject having a large gradation difference from a bright part to a dark part can be photographed by selecting a certain image.

JP-A-9-266545

  However, in the conventional technique as disclosed in Patent Document 1, since the recognition is performed by compositing a region having a good contrast, the photographing condition of the camera is bright as in the case where the vehicle enters the tunnel. When there is a sudden change from an environment to a dark environment, it is inevitable that a delay in recognition occurs.

  The present invention has been made in view of the above circumstances, and when processing a camera image and recognizing the traveling environment of the vehicle, even if the shooting condition of the camera changes suddenly from a bright environment to a dark environment, the recognition is delayed. An object of the present invention is to provide a driving support device capable of avoiding the problem.

  In order to achieve the above object, a driving support device according to the present invention is a driving support device that recognizes a driving environment by processing an image captured by a stereo camera mounted on a vehicle and provides driving support to a driver. Recognizing a tunnel-like structure existing in front of the host vehicle from the captured image, storing a distance between the recognized tunnel-like structure and the host vehicle, and from the host vehicle to the tunnel-like structure And an exposure adjustment unit that adjusts an exposure amount of an image captured by the stereo camera to be appropriate in the tunnel-like structure at a timing determined based on the distance.

  According to the present invention, when a camera image is processed to recognize a traveling environment of a vehicle, a recognition delay can be avoided even if the imaging condition of the camera changes suddenly from a bright environment to a dark environment.

Overall configuration diagram of the driving support device Functional block diagram for shutter speed control Explanation of tunnel recognition Shutter speed switching control flowchart

Embodiments of the present invention will be described below with reference to the drawings.
A driving assistance apparatus 1 shown in FIG. 1 is mounted on a vehicle such as an automobile (not shown), recognizes an external traveling environment, and performs various driving assistances for a driver. The driving support control by the driving support device 1 includes, for example, pre-crash brake control for predicting danger and avoiding collision or reducing collision damage, cruise control with a tracking function in all vehicle speed ranges, There is an alarm control that monitors the wobbling in the lane and the deviation from the lane and warns the driver.

  The driving support device 1 is mainly configured by a stereo camera unit 2 that performs driving environment recognition and driving support control by three-dimensional image recognition, and various controls are performed on the stereo camera unit 2 via a communication bus 10 made of CAN or the like. The unit is connected. The control unit connected to the communication bus 10 is a meter comprising an engine controller 3 for engine control, a transmission controller 4 for transmission control, a vehicle dynamics controller (VDC) 5 for vehicle behavior control, a multi-function display (MFD), and the like. The meter control unit 6 or the like for performing display control.

  The stereo camera unit 2 includes a stereo camera 20 including two cameras as a travel environment recognition sensor. The stereo camera 20 is composed of a pair of (left and right) cameras 20a and 20b each having an image sensor such as a CCD or a CMOS. For example, the two cameras 20a and 20b that are synchronized with each other with variable shutter speeds can transmit light to each other. The object is imaged from different viewpoints by mechanically fixing with a predetermined base line length (optical axis interval) so that the axes are substantially parallel.

  The stereo camera unit 2 including such a stereo camera 20 includes a stereo camera 20, an image processing engine 21 that processes a pair of stereo images captured by the stereo camera 20 and calculates three-dimensional distance information at high speed, and a stereo. A recognition controller 22 that performs object recognition processing such as a white line on the road, a preceding vehicle, and an obstacle on the road using the original image from the camera 20 and the three-dimensional distance information, and each control for driving support based on the recognition result Is configured as a unit integrated with a controller 23 that controls the vehicle, and is installed inside the front window at the upper part of the vehicle interior.

  The image processing engine 21 calculates a shift amount of a corresponding position from a pair of stereo images of the surrounding environment outside the host vehicle imaged by the stereo camera 20 by block matching or the like, and generates a distance image based on the shift amount as distance information. . The image processing engine 21 also has functions of performing shutter speed control of each camera 20a, 20b, geometrical and optical shift correction between both cameras, brightness balance correction, and the like. Is processed at a rate of, for example, 1 frame / 0.1 second (a rate of 1 out of 3 TV images).

  The image processing engine 21 and the recognition controller 22 are connected to each other via the internal bus 2a. The recognition controller 22 performs a grouping process for grouping data within a predetermined threshold on the distance information from the image processing engine 21, and stores the three-dimensional road shape data, sidewall data, Compared with the frame (window) of solid object data, etc., white line data, side walls data such as guardrails and curbs along the road are extracted, and solid objects are two-wheeled vehicles, ordinary vehicles, large vehicles, pedestrians. Categorized into other three-dimensional objects such as utility poles.

  For each recognized data, the respective positions in the coordinate system with the own vehicle as the origin and the longitudinal direction and the width direction of the own vehicle as axes are calculated. In particular, in the vehicle data of two-wheeled vehicles, ordinary vehicles, and large vehicles, the length in the front-rear direction is estimated in advance, for example, 3 m, 4.5 m, 10 m, etc., and the width direction is the center position of the detected width. Is used to calculate the center position of the vehicle.

  Further, in the three-dimensional object data, the relative speed with respect to the own vehicle is calculated from the change in the axial direction of the distance from the own vehicle, and each three-dimensional object is calculated by calculating the relative speed in consideration of the speed of the own vehicle. The speed in each axis direction is calculated. Each information obtained in this way, that is, white line data, side walls data such as guardrails and curbs along the road, and three-dimensional object data (type, distance from own vehicle, center position coordinates, speed, etc.) Therefore, a moving object such as a pedestrian or a light vehicle around the own vehicle and other vehicles traveling on a road connected to the road on which the own vehicle travels is recognized.

  The control controller 23 receives white line data, guardrails existing along the road, side wall data such as curbs, solid object data, and the like from the recognition controller 22. In addition, vehicle information such as the vehicle speed and yaw rate of the host vehicle is input to the controller 23 via the communication bus 10, and based on these information, pre-crash brake control, cruise control with a follow-up function, stagger and lane departure Command the driver for driving support control such as alarm control.

  Here, an image captured by the stereo camera 20 may be captured under a condition where the brightness is different from the surroundings of a bridge-like structure or the like that covers the sky of a tunnel or road. When the host vehicle enters such a tunnel-like structure (hereinafter referred to as “tunnel” as a representative), a slight delay occurs until the shutter speed is controlled to an appropriate shutter speed, and the recognition response delays. Is concerned.

  That is, the shutter speed of the stereo camera 20 is controlled by the image processing engine 21 using a map storing the appropriate shutter speed for each area of the image under auto iris, and the average brightness value of each area of the captured image, etc. The shutter speed is determined by summing up the map values for each area obtained using as a parameter. Therefore, it takes time until the shutter speed is changed after entering the tunnel, which may cause a recognition delay.

  For this reason, as shown in FIG. 2, the recognition controller 22 receives a recognition result from the tunnel recognition unit 22a for recognizing (detecting) the tunnel ahead of the host vehicle from the captured image of the stereo camera, and the recognition result from the tunnel recognition unit 22a. An exposure adjusting unit 22b that adjusts the exposure amount of the captured image by switching the shutter speed of the stereo camera 20; With the functions of the tunnel recognizing unit 22a and the exposure adjusting unit 22b, the recognition controller 22 switches the shutter speed of the stereo camera 20 before entering the tunnel when the presence of the tunnel is recognized in front of the host vehicle, and the captured image immediately after entering the tunnel. By optimizing the exposure amount, recognition delay is avoided.

  In this embodiment, the tunnel recognition unit 22a recognizes a tunnel based on the recognition based on the luminance distribution and distance data in the image. That is, the original image and the distance image obtained by imaging the tunnel entrance have characteristics that the brightness of the central portion is lower than the left and right side portions of the image and that the central portion is farther than the left and right side portions of the image. Accordingly, the tunnel recognizing unit 22a determines that there is a high possibility of being a tunnel entrance when detecting the above-described feature, and compares it with the image pattern of the tunnel entrance stored in advance in the memory, and the brightness of the road surface of the traveling path. A final decision is made using a comparison with

  Specifically, as shown in FIG. 3, when a wall-like object having the same distance is detected in the left and right areas GL and GR in the image, the luminance of the central area GO between the left and right wall-like objects is detected. Check out. And when the brightness | luminance of this area | region GO of the center part is relatively lower than the brightness | luminance of the wall-like object of the right-and-left area | regions GL and GR, it judges that it is highly likely that it is a tunnel entrance, and brightness | luminance is relatively high In such a case, it is determined that the user is in an environment such as a valley of a building.

  Furthermore, when pattern matching with the tunnel entrance image held in advance, comparison with the road surface brightness of the traveling road, etc. are satisfied, when satisfying at least one of the condition that the pattern matches, the condition that the difference from the road surface brightness is equal to or greater than a threshold value, It is finally determined that it is a tunnel entrance. Then, the distance from the host vehicle to the tunnel entrance is obtained and stored in the memory as time-series data corresponding to the progress of the host vehicle.

  Depending on the environmental conditions, it is possible to recognize the tunnel by any one of recognition based on luminance distribution and distance data in the image, recognition for pattern matching, and recognition based on road surface luminance.

  When the tunnel recognition unit 22a recognizes a tunnel existing in the traveling direction, the exposure adjustment unit 22b detects the shutter speed of the stereo camera 20 at a timing before entering the tunnel determined using the distance from the host vehicle to the tunnel entrance. The shutter speed adjustment value is transmitted to the image processing engine 21 so that the normal exposure value is set to a value that provides proper exposure in the tunnel. The shutter speed is adjusted at a timing with the following (A) or (B) as a threshold.

(A) Set distance to tunnel entrance When the distance from the host vehicle to the tunnel entrance is equal to or less than a preset distance, the shutter speed of the stereo camera 20 is changed from a normal value to a value at which proper exposure is performed in the tunnel. adjust.
(B) Set time before entering the tunnel The time to enter the tunnel is calculated from the distance from the own vehicle to the tunnel entrance and the speed of the own vehicle, and when this time becomes less than the set time, the stereo camera 20 The shutter speed is adjusted from a normal value to a value that provides proper exposure in the tunnel.

  In the timing (A), when the set distance is a constant value, the number of frames to be processed before entering the tunnel differs depending on the vehicle speed as compared to the timing (B). Therefore, although depending on the setting of the distance threshold, it is possible to make it equivalent to (B) by changing the set distance according to the vehicle speed. In the following description, it is assumed that the shutter speed is adjusted at the timing according to the set time of (B).

  The shutter speed adjustment value is an instruction value for switching the current shutter speed to a tunnel shutter speed held in advance. For example, the shutter speed of the stereo camera 20 is 1/4000, 1/2000, 1/1000. , 1/500, 1/250, 1/125, 1/50, the shutter speed adjustment value is set so that the shutter speed is switched step by step, with 0 being the reference value for no shutter speed change. , ± 1, ± 2,... The shutter speed for the tunnel is a value set in consideration of the brightness in the typical tunnel obtained by examining the tunnel in each region and the dynamic range of the stereo camera 20, and is stored in the system in advance. Yes.

The set time that is the timing for switching the shutter speed can be set to a constant value according to the processing capacity of the system or the like, but may be varied according to the environmental conditions and the running conditions. At this time, as shown in the following (1) to (3), the switching timing can be appropriately selected.
(1) The current shutter speed is switched to the tunnel shutter speed at once.
(2) The current shutter speed is gradually switched toward the tunnel shutter speed until the host vehicle enters the tunnel.
(3) The current shutter speed and the tunnel shutter speed are switched every predetermined frame until the host vehicle enters the tunnel.

  In the shutter speed switching of (3), a region having good contrast is extracted from images with different shutter speeds until entering the tunnel, and the recognition processing of the driving environment is performed by combining the extracted regions. Is possible.

  After entering the tunnel, the tunnel shutter speed switched according to the adjustment value is maintained until an appropriate shutter speed is obtained under the control of the image processing engine 21. In this case, since the brightness is kept constant in the illuminated tunnel, it is possible to maintain the shutter speed for the tunnel depending on the length of the tunnel.

  In the present embodiment, the shutter speed is switched as exposure amount adjustment, but the exposure amount may be adjusted by changing the iris value of the camera and the gain of an amplifier that amplifies the signal from the image sensor. However, when changing the iris value, it is necessary to switch the shutter speed earlier than the timing of switching the shutter speed in consideration of the response, and therefore, the iris value is gradually switched using the adjustment map. Further, the switching of the shutter speed and the change of the iris value and the gain can be applied during night driving.

  Next, shutter speed switching control by the recognition controller 22 will be described with reference to the flowchart shown in FIG.

  In this shutter speed switching control, first, in the first step S1, recognition processing is performed based on information from the image processing engine 21 that processes the image captured by the stereo camera 20, and whether or not a tunnel exists in front of the host vehicle. Check out. As a result, when the tunnel is not recognized, the present process is skipped and the normal shutter speed control by the image processing engine 21 is performed.

  On the other hand, when the tunnel is recognized in step S1, the process proceeds from step S1 to step S2, and the distance between the host vehicle and the tunnel entrance is acquired. Next, in step S3, a time T until the host vehicle enters the tunnel is obtained from the distance to the tunnel entrance and the speed of the host vehicle, and in step S4, it is checked whether or not the time T is equal to or less than the set time Tset. .

  As a result, when T> Tset, the present process is terminated from step S4, and when T ≦ Tset, the process proceeds from step S4 to step S5, and the shutter speed adjustment value is transmitted to the image processing engine 21, and the stereo camera. The shutter speed of 20 is switched to a shutter speed at which proper exposure is obtained immediately after entering the tunnel.

  As described above, in the present embodiment, when driving assistance by image recognition using the stereo camera 20 is performed, the distance between the tunnel-like structure recognized from the captured image of the stereo camera and the host vehicle is stored. Then, in order to adjust the exposure amount of the image captured by the stereo camera to be appropriate within the tunnel-like structure at a timing determined based on this distance, the exposure amount of the captured image immediately after entering the tunnel is optimized and recognized. Can be avoided.

DESCRIPTION OF SYMBOLS 1 Driving assistance apparatus 2 Stereo camera unit 20 Stereo camera 21 Image processing engine 22 Recognition controller 22a Tunnel recognition part 22b Exposure adjustment part

In order to achieve the above object, a driving support device according to the present invention is a driving support device that recognizes a driving environment by processing an image captured by a stereo camera mounted on a vehicle and provides driving support to a driver. Recognizing a tunnel-like structure existing in front of the host vehicle from the captured image, storing a distance between the recognized tunnel-like structure and the host vehicle, and from the host vehicle to the tunnel-like structure Exposure adjustment that switches the shutter speed of the stereo camera stepwise toward the shutter speed at which the exposure amount of the image captured by the stereo camera is appropriate in the tunnel-like structure at a timing determined based on the distance And a section.

Claims (4)

In a driving support device that recognizes a driving environment by processing an image captured by a stereo camera mounted on a vehicle and performs driving support for a driver,
A recognition unit for recognizing a tunnel-like structure existing in front of the host vehicle from the captured image of the stereo camera, and storing a distance between the recognized tunnel-like structure and the host vehicle;
An exposure adjustment unit that adjusts an exposure amount of an image captured by the stereo camera to be appropriate in the tunnel-like structure at a timing determined based on a distance from the host vehicle to the tunnel-like structure. A driving support device characterized by comprising.   The exposure adjusting unit calculates a time until the host vehicle enters the tunnel-shaped structure from the distance from the host vehicle to the tunnel-shaped structure and the vehicle speed of the host vehicle, and this time is set time. The driving support device according to claim 1, wherein an exposure amount of an image captured by the stereo camera is adjusted at a timing of   The driving support apparatus according to claim 1, wherein the exposure adjustment unit adjusts an exposure amount of an image captured by the stereo camera at a timing when a distance from the own vehicle to the tunnel-like structure becomes a set distance. .   The driving support apparatus according to claim 1, wherein the exposure adjustment unit adjusts the shutter speed of the stereo camera to be a preset shutter speed.

Images