JP2012221134A - Driving support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a driving support device capable of surely starting to record video, even when an own vehicle is about to collide with a preceding vehicle.SOLUTION: The driving support device comprises: a video acquiring section 10 for acquiring outside video; an image recognizing section 20 for recognizing lighting of stop lamps or a turn signal of the preceding vehicle from the video acquired from the video acquiring section, and outputting a recognized result; a video recording section 30 for recording the acquired video; and a control section 40 for starting to record the video on the video recording section when the lighting of lights of the preceding vehicle is recognized on the basis of the recognized result from the image recognizing section.

Description

  The present invention relates to a driving support device that keeps a record of external conditions during driving when driving a vehicle.

  Conventionally, as a drive recorder, an apparatus for recording an image in front of the vehicle immediately before an accident has been developed. As a condition for starting the video, a change in acceleration or jerk (jerk) is detected. If the change is rapid, it is determined that there is an accident or a near-miss situation, and an external video is recorded. The recorded result can be taken out of the apparatus, and can be used for investigation of the cause of the accident and analysis of the cause of the accident based on the taken images and images.

  Furthermore, there is a need to view not only the recorded video before and after the accident but also when the danger of a collision increases (a scene of a near-miss), and a drive recorder that starts video recording before the collision has been developed. Yes. It is known that the recorded video can be used for driver education as an example. For example, it is a device that can detect a front obstacle and record a video using a stereo camera as described in Patent Document 1.

  As a condition for starting recording, if only a situation of rapid deceleration is taken as a condition, recording is not started in the following example of a scene of a rear-end collision.

  If the driver of the vehicle is looking forward, such as when the driver of the vehicle is braking and starting to decelerate, the vehicle will collide with the preceding vehicle with almost no deceleration. There is a case. At that time, recording is not started by a method that relies solely on own vehicle information such as the deceleration of the own vehicle.

  As a method for recognizing the situation of the preceding vehicle, there are external environment recognition devices using millimeter waves and cameras. Using the external environment recognition device, the time to collision (TTC: Time To Collision) is calculated (= relative distance / relative speed) and compared with a threshold value to give an alarm or apply a brake. It is used to reduce collision damage. By measuring the TTC by a method using this external environment recognition device, it is possible to start recording before the collision and take a near-miss video. In the method using the external environment recognition device, the state of the preceding vehicle can be recorded from the recording start time.

  However, if the stop lamp of the preceding vehicle has already been lit at the start of recording, the video after recording can be confirmed, but the behavior of the preceding vehicle and the behavior of the host vehicle after the stop lamp is lit before recording. Is unknown. That is, by introducing a method for starting recording from the lighting of the stop lamp of the preceding vehicle, which could not be obtained in the past, it can be used for near-miss or accident cause investigation. Furthermore, in order to lower the recording frequency, the recording start condition may be changed by taking the AND condition between the stop lamp lighting and the TTC.

  In addition, the technique itself which detects lighting of the stop lamp of a preceding vehicle is described in patent document 2 and patent document 3.

JP 2010-86362 A JP-A-8-335300 JP-A-10-236249

  The problem to be solved by the present invention is to provide a driving support device that can reliably start video recording even when it is likely to collide with a preceding vehicle.

  In order to solve the above problem, the driving support device of the present invention recognizes the lighting of a stop lamp of a preceding vehicle or a turn signal from a video acquisition unit that acquires an external video, and a video acquired from the video acquisition unit, Based on the recognition result from the image recognition unit for outputting the recognition result, the video recording unit for recording the acquired video, and the recognition result from the image recognition unit, if the lighting of the preceding vehicle is recognized, the video to the video recording unit And a control unit for starting recording.

  The present invention can provide a driving support device that can reliably start video recording even when it is likely to collide with a preceding vehicle.

It is a figure which shows the structural example of the driving assistance device which concerns on this invention. It is a figure which shows the operation | movement flow of the driving assistance device which concerns on this invention. It is a figure explaining the judgment conditions of the control part of the driving assistance device of the present invention. It is a figure which shows the function example of the control part of the driving assistance device of this invention. It is a figure which shows the other structural example of the driving assistance device which concerns on this invention. It is a figure which shows the other structural example of the driving assistance device which concerns on this invention. It is a figure explaining the other determination conditions of the control part of the driving assistance device of the present invention. It is a figure which shows the other operation | movement flow of the driving assistance device which concerns on this invention.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 shows a configuration diagram of a driving support apparatus according to the present invention. The video acquisition unit 10 is a camera unit that is an imaging device that acquires external video.

  The video acquisition unit 10 is a monocular camera having one image sensor or a stereo camera having two left and right image sensors. The image recognizing unit 20 performs image recognition based on the video acquired by the video acquiring unit 10, recognizes lighting of an obstacle or a preceding vehicle (stop lamp, hazard lamp, turn signal), and outputs a recognition result. The image recognition unit 20 may measure the inter-vehicle distance from the preceding vehicle and measure the time until the vehicle collides with the preceding vehicle. The video recording unit 30 records the video acquired by the video acquisition unit 10 in various memories and media. The control unit 40 acquires the recognition result of the image recognition unit 20, determines whether or not it is a stop lamp, and records the video in the video recording unit 30 if it is determined that the stop lamp is turned on. That is, when the lighting of the preceding vehicle is recognized based on the recognition result from the image recognition unit 20, the control unit 40 starts recording the video on the video recording unit 30.

  The communication unit 50 performs communication for inputting vehicle information such as vehicle speed, acceleration, and yaw rate to the control unit 40 via the vehicle network 60, and receives a recording start signal from the control unit 40, which is the driving support device of the present invention. Obstacle recognition information and the like are output via the vehicle network 60. The acceleration detection unit 70 notifies the control unit 40 when sudden deceleration occurs. If an equivalent signal is obtained from the vehicle network 60, it can be omitted.

  Next, an operation flow of the control unit 40 which is a driving support device will be described with reference to FIG.

  In S210, the video acquisition unit 10 acquires an external video.

  Next, in S220, the image recognition unit 20 recognizes the stop lamp of the preceding vehicle based on the acquired external video.

  Next, in S230, the control unit 40 performs condition determination based on the recognition result in the image recognition unit 20. If the stop lamp is recognized, the determination is Yes, the process proceeds to S240, and the video is stored in the memory. There are various video storage methods, and there is a method of storing in the video recording unit 30 using a ring buffer so as to suppress the actual data transfer operation.

  Next, in S250, the memory is locked so that the video stored in the video recording unit 30 is not overwritten.

  If the determination in S230 is No, the process returns to the first S210 and the video acquisition is continued.

  FIG. 3 shows an example of determination conditions of the control unit 40 in S230.

  Determination condition 1 is a conventional technique, and the image recognition unit 20 calculates a time TTC at which a collision with a preceding vehicle is expected from the input relative distance and relative speed, and the TTC output as a recognition result is determined in advance. Recording is started when it is determined that it is smaller than the first threshold (short: about 1 second).

  In this embodiment, as the determination condition 2, the image recognition unit 20 starts recording when a stop lamp is lit or a turn signal is recognized from the acquired video. Alternatively, as the determination condition 3, when the image recognition unit 20 recognizes that the stop lamp is turned on or the turn signal is recognized, and the TTC has a predetermined second threshold (long: several seconds or more) (first threshold < When the value is smaller than the second threshold value), the recording is started.

  By determining the condition in this way, even when it is likely to collide with the preceding vehicle, the recording can be surely started from the brake start time of the preceding vehicle.

  Next, FIG. 4 shows an example of function expansion of the control unit 40.

  As a configuration, in addition to the single function of the drive recorder that records video in the video recording unit 30 as shown in the configuration diagram of FIG. 1, the control unit 40 can have an additional function. For example, as shown in FIG. 4, an obstacle recognition function (function 2), an inter-vehicle control function (function 3) such as ACC control, and a lane departure prevention control function (function 4) can be combined. The combination 1 is configured to output the obstacle recognition result to the vehicle network in addition to the function of the drive recorder. In addition, the combination 2 has a control function such as an inter-vehicle distance control and a lane departure function added to the control unit 40 and has a combination.

  As described above, according to the present invention, the behavior of the preceding vehicle and the host vehicle in the forward direction triggered by the lighting of the preceding vehicle lamp (stop lamp, hazard lamp, turn signal) is recorded, which is effective for analyzing the cause of the accident. . The recorded video information can be taken out and used as evidence, and can also be used as teaching materials for driver education.

  Next, FIG. 5 shows another configuration diagram of the driving support apparatus according to the present invention. The present embodiment shows recording on the video recording unit 30 when an in-vehicle stereo camera is used as the video acquisition unit.

  The configuration of FIG. 5 is that the video acquisition unit 100 is a stereo camera, acquires a stereo image, the image recognition unit 200 performs image recognition using the stereo image acquired by the video acquisition unit 100, and the video recording unit 300. That the video to be stored in the left and right camera images acquired by the video acquisition unit 100 can be recorded, the vehicle can be controlled by the control unit 400 based on the recognition result recognized by the image recognition unit 200, and the acceleration detection unit The absence of 70 is different from Example 1 (FIG. 1). The rest is the same as FIG.

  The image recording unit 300 can record the left and right stereo images, and the distance image with the obstacle such as the preceding vehicle recognized or calculated by the image recognition unit 200 or the recognition result of the obstacle in the front. It can be recorded in synchronization with external video.

  Further, the communication unit 50 can obtain the vehicle speed signal and acceleration signal to the control unit 400 via the vehicle network 60 and omit the acceleration detection unit 70 of FIG. 1 to reduce the cost.

  Next, FIG. 6 shows another configuration diagram of the driving support apparatus according to the present invention. In the present embodiment, a configuration in which a video, a horn, and an alarm sound are recognized and a video is recorded in the video recording unit 30 is shown.

  In the configuration of FIG. 6, a voice recognition unit 80 is added to the configuration of FIG. 1 of the first embodiment. In the voice recognition unit 80, a vehicle horn (alarm sound), an alarm sound output from an in-vehicle device, The voice is recognized. Speech recognition is well known and will be omitted.

  FIG. 7 shows an example of the determination conditions of the control unit 40 in this embodiment. An example of the determination condition when voice recognition is added will be described.

  Determination conditions 1 to 3 are the same as the determination conditions of FIG.

  The determination condition 4 is that the voice recognition unit 80 recognizes a sound such as an alarm sound, and starts the recording by using the recognition as a trigger.

  The determination condition 5 is that the image recognition unit 20 recognizes the lighting of the stop lamp or the turn signal and the voice recognition unit 80 recognizes the sound such as an alarm sound as a trigger to start recording.

  The determination condition 6 is that the image recognition unit 20 recognizes the lighting of the stop lamp or the turn signal, the voice recognition unit 80 recognizes the sound such as an alarm sound, and the image recognition unit 20 collides with the preceding vehicle. Is calculated, and when it is determined that the calculated TTC is smaller than a predetermined second threshold (long: several seconds or more), recording is started.

  Each of these combinations is described as an AND condition, but an OR condition may be incorporated.

  Next, FIG. 8 shows a processing flow for automatically determining the recording start of the video recording unit 30 of the control unit 40 as the fourth embodiment.

  First, in S810, the image recognition unit 20 acquires information (relative distance and relative speed) necessary for calculating the time TTC at which a collision with the preceding vehicle is expected, and calculates the TTC.

  Next, in S820, the image recognition unit 20 recognizes lighting (stop lamp or turn signal) from the external video acquired by the video acquisition unit 10. In step S830, the control unit 40 determines whether the TTC input from the image recognition unit 20 is smaller than a predetermined threshold value 1 (first threshold value). When the determination is Yes (when TTC <threshold 1), the situation is just before the collision, so that recording to the video recording unit 30 is started immediately in S860. When the determination is No (when TTC> threshold 1), the process proceeds to S840. In step S840, the control unit 40 determines whether the TTC input from the image recognition unit 20 is smaller than a predetermined threshold value 2 (second threshold value) greater than the threshold value 1. If Yes (threshold 1 <TTC <threshold 2), the process proceeds to S850. In S850, when the image recognition unit 20 recognizes and recognizes lighting of the vehicle (stop lamp or turn signal), that is, in the case of Yes determination, recording to the video recording unit 30 is started in S860. When the determination is No in S840 or S850 (when TTC> threshold 2), the process proceeds to S870, and the storing operation of the video in the video recording unit 30 is terminated.

  Here, the conditions for starting and ending the recording are set easily, but the recording can be continued for a certain period of time after the recording starts, or the recording can be ended under other conditions, A setting method such as continuing recording until time elapses can be easily configured using state transition and a timer.

10, 100 Video acquisition unit 20, 200 Image recognition unit 30, 300 Video recording unit 40, 400 Control unit 50 Communication unit 60 Vehicle network 70 Acceleration detection unit 80 Voice recognition unit

Claims (7)

A video acquisition unit for acquiring external video;
An image recognition unit that recognizes a stop lamp of a preceding vehicle from the video acquired from the video acquisition unit or lighting of a turn signal, and outputs a recognition result;
A video recording unit for recording the acquired video;
And a control unit that starts recording the video in the video recording unit when the lighting of a preceding vehicle is recognized based on the recognition result from the image recognition unit. The driving support device according to claim 1,
The image recognition unit calculates a time TTC at which a collision with a preceding vehicle is expected from the input relative distance and relative speed, and outputs the calculated TTC as the recognition result;
The control unit is a driving support device that starts recording the video to the video recording unit when the lighting of a preceding vehicle is recognized and the TTC is smaller than a predetermined threshold. The driving support device according to claim 1,
The video acquisition unit is a driving support device that includes two image sensors and is a stereo camera that acquires a stereo video. The driving support device according to claim 1,
A driving support device having a voice recognition unit that recognizes a warning sound of a host vehicle or another vehicle or a voice of a pedestrian and outputs a recognition result. The driving support device according to claim 4,
The control unit is configured to detect the video when the lighting of a preceding vehicle is recognized and an alarm sound or voice is recognized based on the recognition result from the image recognition unit and the recognition result from the voice recognition unit. A driving support device that starts recording the video on a recording unit. The driving support device according to claim 4,
The image recognition unit calculates a time TTC at which a collision with a preceding vehicle is expected from the input relative distance and relative speed, and outputs the calculated TTC as the recognition result;
The control unit is configured to recognize the lighting of a preceding vehicle based on the recognition result from the image recognition unit and the recognition result from the voice recognition unit, and when the TTC is smaller than a predetermined threshold, and A driving support device that starts recording the video to the video recording unit when an alarm sound or voice is recognized. The driving support device according to claim 1,
The image recognition unit calculates a time TTC at which a collision with a preceding vehicle is expected from the input relative distance and relative speed, and outputs the calculated TTC as the recognition result;
The control unit recognizes that the light of the preceding vehicle is turned on when the TTC is greater than a predetermined first threshold and less than a predetermined second threshold greater than the first threshold. A driving support device that starts recording the video on the video recording unit.

Images