JPS61135842A - Substance recognition system for a car - Google Patents

Abstract

PURPOSE:To enable an accurate distance recognition to a substance from a car, by changing substance recognition area according to the running condition such as the car speed and the steering angle. CONSTITUTION:A forward distance equal to the safety interval distance responding to a car speed, read by the speed sensor 1 is a focusing distance. Here, focus distance is calculated comparing the forward substance recognition distance and the focusing distance. Then zooming regulation is carried out to seek for the focus distance TV cameras 4 and 5 have decided to get sampling points. The distance between the sampling points is decided according to the substance between the sampling points is decided according to the substance recognition distance. Moreover, when turning a curve, the zooming is carried out to expand the field of view according to the steering angle. The data of interval between the sampling points is delivered to the delaying circuit 10 through a microcomputer 3. Thus, substance recognition in a given extent around the center of the forward recognition point is being possible.

Description

DETAILED DESCRIPTION OF THE INVENTION (A) Field of Industrial Application The present invention relates to an object recognition device for a vehicle that can accurately recognize objects around a vehicle even when the running condition of the vehicle changes.

(b) Prior Art Recently, unmanned vehicles that drive around while recognizing and making judgments about the situation in front of and around the vehicle are being considered.

Conventional cars of this type use two television cameras placed above and below to recognize vehicles and obstacles in front, and use the points where the brightness and darkness of the image changes to detect objects located at the intersection of the optical axes of the cameras. It takes advantage of the fact that only the images of the two television cameras overlap when the image planes of the two television cameras are compared (for example, Japanese Patent Publication No. 48-2979).

If such an object recognition device is installed, it will be possible to recognize vehicles or obstacles in front or behind when driving on gentle curves or at low speeds, which is good for drivers and autopilots, but it is better to use a fixed-focus camera. Therefore, when trying to see a relatively far distance, such as when driving at high speed, the difference between the images taken by the two television cameras becomes small, making it susceptible to noise and the like, making it impossible to detect obstacles.

(c) Object and structure of the invention The present invention has been made in view of the above points, and aims to enable accurate recognition of objects around a vehicle. The recognition area is configured to change based on driving conditions such as steering angle and steering angle.

FIG. 1 shows the overall configuration of an object recognition device for a vehicle according to the present invention, which detects the running state of the vehicle using a vehicle speed sensor and a steering angle sensor, and locates the object recognition position from the vehicle based on the detected running state. The recognition area control means calculates the distance to
The object recognition area is changed based on the calculated distance.

) Example The present invention will be explained below based on the drawings.

FIG. 2 is a block diagram of an embodiment of the vehicle object recognition device according to the present invention, in which 1 is a vehicle speed sensor that detects the speed of the vehicle, 2 is a steering angle sensor that detects the steering angle of the vehicle, and 3 is a CPU. , a λL microcomputer consisting of memory, etc.; 4 and 5 a television camera equipped with an electric zoom mechanism and a focusing mechanism; 6.7 a differentiation circuit for differentiating the video signal output from the television camera 4.5; 8. 9 is a waveform shaping circuit that shapes the output of the differentiating circuits 6 and 7; 10 is a delay circuit that delays one differential output by a variable amount than the other differential output; 11 is a delayed differential output and an undelayed differential output. 12 is a monitor television that monitors images of objects on the road captured by a television camera 4.5.

The basic operation of object recognition by the object recognition apparatus described above is not the gist of the present invention and is described in detail in the aforementioned Japanese Patent Publication No. 48-2979, so a brief description thereof will be provided.

A three-dimensional target space is captured by two TV cameras 4.5
A video signal as a dimensional time function is extracted and differentiated by a differentiating circuit 6.7 to obtain a pulse train signal. The pulse train signal corresponds to a place where brightness changes drastically. After shaping the waveforms of these series of pulse train signals in the waveform shaping circuit 8.9, one of them is delayed by a certain amount of time in the delay circuit 10, and then inputted into the AND circuit 11 to perform the logical product. This corresponds to brightness change points distributed in the height direction at a distance X. By arbitrarily changing the amount of delay in the delay circuit 10, it is possible to extract changes in light and dark spots in the direction of the distance at a distance corresponding to the amount of delay, which also indicates that a three-dimensional object exists at that position. means. If there is no three-dimensional object on the road, there will be no change in bright and dark spots in the height direction, so the output of the MΦ circuit 11 will be only one pulse.

Next, the operation according to the present invention will be explained with reference to the flowchart of FIG.

When g-identifying an object, initial settings are made and the object recognition distance X (for example, 120 m) is set (F-
1). Next, the vehicle speed V is determined by the vehicle speed sensor 1.

and read the steering angle θ by the steering angle sensor 2 (
F-2). Since the safe inter-vehicle distance with respect to vehicle speed has a relationship as shown in FIG. 4, the safe inter-vehicle distance l with respect to the read vehicle speed v1 is determined. The distance ahead corresponding to this safe inter-vehicle distance becomes the focus distance P (F-3). Here, compare the object recognition distance in front of the vehicle and the focus distance #p.
) If x, set X to P and calculate the focal length f using the following formula (F-5); if P≦X, use P as is (
F-6).

f=”tan-3θ ll where d is the angle of view of the lens, WH is the shooting width at the recognition position,
θ is a steering angle, and a is a constant. As can be seen from this equation, as the steering angle θ increases, the focal length f decreases.

The zoom is adjusted so that the focal length of the television camera 4.5 becomes the focal length f thus determined, and the sampling point is determined (F-7). The sampling point interval is determined according to the object recognition distance X. For example, (1) When the distance X to the point t1t that you want to recognize is less than 50 m. Taking X = 10 m as an example, as shown in Figure 5, from L1 = 5 m to 5 = 20 m, The area is recognized by dividing it into 16 band-shaped areas at intervals of 1 m.

In the figure, G indicates a guardrail, and the white part indicates the outside of the field of view.

If we express the above in a formula, (However, n is the number of divisions on the screen.)

(2) When the distance X to the position you want to recognize is 50 m or more. Taking X = 50 m as an example, as shown in Fig. 6, from = 25 m, 1. = 100 m
From Wl to L, = 25 m to La” 50 m
'Recognize in J.

Expressing the above as a formula (n is expressed as the number of divisions on the screen), 5 Li = Li - 1 + □ where i > -)
2. Also, the width of the band-shaped area is adjusted according to the steering angle θ when making a curve, and zooming is performed to widen the field of view.

Data Li having such a sampling point interval is sent from the microcomputer 3 to the delay circuit 10, and the output of the waveform shaping circuit 8 is delayed by a delay time corresponding to this li. In this way, it becomes possible to recognize objects within a predetermined range centered on the recognition position in front of the vehicle (F-8).

The image of the recognition area is displayed on the monitor television 12 (F-
9), the distance from the vehicle to the object is calculated (F-10
).

If the distance information to the object obtained in this manner is used as a new focal length P to recalculate the focal length f and the object is recognized, accurate recognition will always be possible.

(E) Detailed Description of the Invention As described above, the present invention provides an object recognition device for a vehicle that recognizes objects around the vehicle from video information around the vehicle captured by an imaging means installed in the vehicle. Since the recognition area is configured to change based on driving conditions such as the steering angle, it is possible to accurately recognize objects such as vehicles in front and road fixtures such as guardrails or traffic signs when driving at high speeds or when cornering. It is possible to accurately determine the distance, which is preferable for safety reasons.

[Brief explanation of drawings]

FIG. 1 is an overall configuration diagram of a vehicle object recognition device according to the present invention, FIG. 2 is a block diagram of an embodiment of a vehicle object recognition device according to the present invention, and FIG. 3 is a recognition area change control according to the present invention. Flowchart showing the operation, Figure 4 is a characteristic diagram showing the relationship between vehicle speed and safe inter-vehicle distance, Figures 5 and 6 are
The figures are two different examples of image recognition areas according to the invention. 1...Vehicle speed sensor, 2...Steering angle sensor, 3...
Microcomputer, 4.5...TV camera, 6
．． 7... Differential circuit, 8.9... Waveform shaping circuit, 10
...Delay circuit, 11... Melon circuit No. 1 @ Figure 4 Vehicle speed W No. 21! I No. 311

Claims (1)

[Claims] A vehicle object recognition device that recognizes objects around a vehicle from video information around the vehicle captured by an imaging means installed in the vehicle includes a detection means for detecting a running state of the vehicle, and a running state detected by the detection means. 1. A vehicle object recognition device, comprising: a recognition area control means for changing a recognition area based on.