JPH0610839B2 - Warning lamp recognition device for the vehicle ahead - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial application] The present invention monitors a braking state of a vehicle traveling ahead, particularly when driving a vehicle, so that an inter-vehicle distance and the like can be kept safe. The present invention relates to a device for recognizing an alarm lamp of a vehicle ahead that calls the driver's attention.

[Background Art] Various safety devices have been developed for use in driving an automobile, and, for example, a drowsy driving preventing means, a backward visual recognition device, and the like are considered. But,
These safety devices are not sufficiently satisfactory, and in particular they cannot recognize the state of vehicles traveling in front, such as the inter-vehicle distance and the changing state of this inter-vehicle distance.

For example, if there is a large difference in the traveling speed between the vehicle traveling ahead and your own vehicle and you are rapidly approaching the vehicle in front, or if the vehicle traveling ahead is turning right or left at the direction indicator lamp, When the notification is given, the driver must accurately recognize the state and perform the driving operation.
Then, by adequately notifying the driver of such a state and calling attention to the driver, a sufficient purpose of safe driving is achieved.

[Problems to be Solved by the Invention] The present invention has been made in view of the above points, and recognizes the state of the vehicle in front by recognizing the lighting state of the braking lamp direction indicator lamp of the vehicle traveling in front. An object of the present invention is to provide a recognition device for an alarm lamp of a vehicle ahead, which notifies the driver and effectively prevents the occurrence of an accident or the like due to the driver's carelessness.

[Means for Solving Problems] That is, an alarm lamp recognition device for a front vehicle according to the present invention includes a color image pickup device for photographing the front of the vehicle, and a video signal of a color image picked up by this image pickup device. From this, for example, a color image signal of a color corresponding to the color of the braking lamp is extracted as a feature. Then, based on the color image signal from which the characteristics have been extracted, the lighting state of, for example, the braking lamp of the vehicle traveling ahead is recognized. In this case, the image pattern of the braking lamp changes in accordance with the change in the distance to the vehicle in front, and the change state of the inter-vehicle distance is calculated by the pattern change over time, and the danger can be predicted. The driver is informed.

[Operation] In the device configured as described above, for example, when the pattern of the braking lamp of the front vehicle is recognized, when the inter-vehicle distance from the front vehicle becomes shorter, the color of the braking lamp is changed. The pattern of the extracted feature image changes. Specifically, the braking lamp is represented by a pair of red points, but if the inter-vehicle distance changes so as to become shorter, the interval between the pair of red points will change over time. As the distance increases, it becomes possible to determine the speed difference between the vehicles from the changing state of the interval, and the risk of a rear-end collision can be predicted. Therefore, if the driver is notified by sound, a lamp or the like in a state where such a danger can be predicted, the driver can effectively cope with this.

[Embodiment of the Invention] An embodiment of the present invention will be described below with reference to the drawings.
FIG. 1 shows its configuration, which includes a color television camera 11 as a color image pickup device. The television camera 11 is attached and set in front of the vehicle 10 as shown in FIG. 2, for example, and is set so that the rear of the vehicle 10a traveling in front can be effectively photographed.

A color image (video) signal based on the color image captured by the television camera 11 is supplied to the decoder 12.
The color image signals of R (red), G (green) and B (blue) are output from the decoder 12.

The R, G, B image signals from the decoder 12 are sent to the image signal processing unit 13. The image signal processing unit 13 is an image of a vehicle that is driving in front of an image taken by the television camera 11 on the basis of the image signal. The vehicle-to-vehicle distance and the relative speed and the like are recognized, and the output signal from the signal processing unit 13 is sent to the determination unit 14.

The determination unit 14 is further supplied with a vehicle speed signal from the vehicle speed sensor 15 and information on the operation state of the brake from the brake pedal 16, and issues a notification command to the notification unit 17 based on these information. , It is to determine whether or not to notify the driver.

FIG. 3 shows a flow of operations of the apparatus configured as described above, particularly in the image signal processing unit 13. First, in response to turning on of the ignition switch of this vehicle, step 10 is performed.
Initially set to 1. That is, the scanning portion of the image to be shot on the screen is set, and the characteristic extraction conditions in color are set. These setting conditions and the like are stored in a storage means such as a ROM. Then, in step 102, the R, G, B color image data output from the decoder 12 is fetched.

Here, the image signal processing unit 13 is configured as shown in FIG. That is, the R, G, and B color image signals from the decoder 12 are input to the feature extraction unit 131. The feature extraction unit 131 extracts the red feature of the braking lamp and the yellow feature of the direction indicator lamp, and extracts only the information regarding the object to be recognized by binarizing the input image data, that is, the brake lamp and the direction indicator lamp. To Specifically, since the red emission color is a state in which the value of red (R) is twice or more that of the other green (G) and blue (B), as when the braking lamp is lit, The conditional expression for extracting the characteristics of the image of the braking lamp is R> 2G, and R> 2B ... (1). In the case of the direction indicator lamp, the emission color is yellow, and the conditional expressions thereof are R> 2B, G> 2B ... (2).

That is, after the image data is input in step 102, the above feature extraction is executed in step 103, and the feature extraction data is stored in the memory 132 as shown in step 104. Such a data storing operation, that is, this processing routine is executed, for example, every 0.5 seconds.

Next, at step 105, the image data stored in the memory 132 is sent to the recognition unit 133, and it is determined whether or not the image data stored in this memory 132 is for a braking lamp. This judgment is made by the recognition unit 133.
As a criterion for the determination in the recognition unit 133, for example, as shown in FIG. 5, are there two red color point images 21a and 21b of the same height within the set range 20 corresponding to the road on the screen? If such a red image of the same height is confirmed, it is determined that this is a braking lamp, and the process proceeds to step 106. This step 106
Then, the image data determined as the above braking lamp is calculated by the calculation unit.
This calculation unit 134 calculates the inter-vehicle distance to the vehicle ahead in which the recognized braking lamp is set and the relative speed to the own vehicle, as shown in step 107. .

Specifically, the recognition operation is executed every 0.05 seconds, and the image data recognized as the braking lamp is input to the calculation unit 134 every 0.05 seconds. Then, in the calculation unit 134, the two images 21a and
Calculate the distance r between 21b. And this distance r
Calculate the following distance from the vehicle ahead.

For example, if the focal length of the television camera 11 is f, the distance from the lens of the television camera 11 to the braking lamp of the vehicle ahead is Z, and the magnification of the camera 11 is β, then β = f / Z .... The relationship of (3) is established. When the distance between the braking lamps is R, the following relationship holds when β is “1”.

β = r / R …………………………………………………………………………………………………………………………………………………………………………………………………… (4)
Is calculated by the following formula.

Z / fR / r …………………… (5) The distance between the braking lights varies depending on the vehicle model,
The difference is small, and the inter-vehicle distance can be roughly calculated by the above equation (5).

Further, as described above, the image data is input to the calculation unit 134 every 0.05 seconds.
0.05 seconds after the image data shown in the figure is input, the image data as shown in FIG. In this case, when the inter-vehicle distance to the vehicle in front changes, the distance r2 between the red images 21a and 21b
Will change. Specifically, when the inter-vehicle distance becomes small, the distance r2 is the distance r of the image shown in FIG. 5 photographed 0.05 seconds before that.
Get bigger. When the inter-vehicle distance obtained based on the image data obtained after 0.05 is Z1, the relative speed V with respect to the preceding vehicle is obtained by the following equation.

V = (Z-Z1) /0.05 (6) That is, the image signal processing unit 13 calculates the inter-vehicle distance and the relative speed with the preceding vehicle as described above, and outputs the calculation result. Therefore, the calculation result is sent to the judgment unit 14 as shown in step 108. The determination unit 14 is supplied with the above-mentioned inter-vehicle distance, the relative speed, and other calculation results, as well as the absolute speed from the vehicle speed sensor 15 and each information indicating whether or not the braking operation from the brake pedal 16 is being performed. ing. Then, in step 109, it is determined whether or not to notify the driver based on various data supplied to the determination unit 14.

For example, when the time until a rear-end collision with a vehicle ahead is predicted from the changing state of the inter-vehicle distance data and the relative speed at every 0.05 seconds, and the braking operation is not executed in a state where the rear-side collision is predicted within 15 seconds, for example. In step S110, it is determined that the driver is notified, and the sound of the buzzer or the like is used to call the driver's attention. In this case, if the buzzer sound is controlled to increase as the predicted time until a rear-end collision decreases, a more effective alarm will be executed.

If it is determined in step 105 that the image data stored in the memory 132 is not the braking lamp, the process proceeds to step 111. In this step 111, it is determined whether or not the image data stored in the feature-extracted memory 132 is an image of a direction signal (turn signal) lamp. It is executed depending on whether or not it blinks at equal time intervals of 0.5 seconds.

Next, at step 112, the relative speed with respect to the vehicle ahead is calculated, and this calculation is executed based on the area s of the light emitter, for example. That is, when the area of the light-emitting body by the actual direction indicator lamp is S and the magnification is "1", the relative speed Z is calculated by the following formula.

Z = fS / s (7) In step 113, the calculated relative speed is, for example, 2
The judgment is made to notify the driver when the speed is 0 km / h or more, and the driver is notified in step 110 according to the judgment result. In this case, for example, when a braking operation is performed, it is effective to stop the notification operation.

It should be noted that the color image capturing means for capturing the front of the vehicle only needs to be capable of capturing the front of the vehicle, and is not limited to the example shown in FIG. Also, the mounting angle of the image pickup device is made variable so that the image pickup direction of the image pickup device is automatically adjusted according to, for example, the steering angle so that the vehicle traveling in front is always set in the image pickup screen. You may do it. Further, the rear-end collision time prediction with the vehicle in front may be simply calculated from the inter-vehicle distance and the relative speed as described in the embodiment, but, for example, 0.
It may be calculated by curve approximation from the change of the relative speed every 05 seconds. In this case, the greater the absolute speed, the longer the braking distance. Therefore, it is effective to advance the notification time as the absolute speed increases. Further, the means for notifying the driver is not limited to the buzzer, but may be a voice that is formed by voice synthesis, or a light-using device such as a lamp may be used in combination.

[Effects of the Invention] As described above, according to the recognition device of the present invention, the traveling state of the vehicle traveling in front is constantly monitored, and particularly, the vehicle is stopped or decelerated. In this case, by recognizing the braking lamp, this can be surely detected. In addition, even when the vehicle ahead is about to bend, it can be automatically recognized by identifying the direction indicator lamp. A great effect will be exerted in order to carry out safe driving such as prevention of accidents.

[Brief description of drawings]

FIG. 1 is a configuration diagram illustrating an alarm lamp recognition device according to an embodiment of the present invention, FIG. 2 is a diagram illustrating a state of a vehicle to which the device is attached, and FIG. 3 is a flow of operation of the device. 4 is a diagram showing the configuration of the image signal processing unit of the above apparatus, and FIGS. 5 and 6 are diagrams showing the states of the recognition images obtained by the above apparatus. 10 …… vehicle, 11 …… color television camera, 12 ……
Decoder, 13 ... Image signal processing unit, 14 ... Judgment unit, 15 ...
… Vehicle speed sensor, 16 …… Brake pedal, 17 …… Notification section.

Claims (1)

[Claims] 1. A color image pickup device set to photograph the front of a vehicle, a feature extraction means for extracting a color image signal of a color specified from a color video signal photographed by the image pickup device, and the feature extraction. Based on the color image signal extracted by the means, a recognizing means for determining whether or not the image is a warning lamp of the vehicle, and a distance to the vehicle in front due to a change with time of the recognized color image signal. In addition, a means for calculating the relative speed and an alarm generating means driven based on the result calculated by this means are provided, and the alarm generating means notifies the driver of the proximity of the vehicle in front. Characteristic vehicle alarm lamp recognition device.