JPS639813A - Detection system for vehicle-to-vehicle distance - Google Patents

Abstract

PURPOSE:To judge an abnormal approach by processing image data obtained by picking up an image of an object and finding the apparent vehicle width of the object on a screen, and displaying variation in the distance to a preceding vehicle with the direction of variation in the vehicle width with time. CONSTITUTION:An image pickup device 1 is installed on a vehicle 2 at the place where its front visual field is obtained sufficiently without any hindrance to driving. The video signal of the device 1 is A/D-converted 3, inputted to a processor 4, and stored in its memory 5. This processor 4 computes the width of a preceding vehicle by its computer 6 from data stored in the memory 5 at specific intervals of time and compares it with the vehicle width obtained by the last measurement to find the direction and speed of the variation in the vehicle width. Then, a signal indicating the direction of the increase or decrease in the vehicle width and an abnormal approach warning signal indicating the width exceeds a specific value in the increasing direction are outputted. Then, a vehicle interval distance meter 7 indicates whether the preceding vehicle is moving close or away on the basis of the arithmetic result of the computer 6 and generates a warning by displaying 'Reduce the speed' an making a buzzer sound when the abnormal approach warning signal is generated.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an inter-vehicle distance measuring method for measuring the distance to a preceding vehicle, and in particular, a method for determining the apparent width of the preceding vehicle from image data. This relates to an inter-vehicle distance detection method that determines the direction of change in the inter-vehicle distance from the change and generates a warning in the event of an abnormal approach.

[Conventional technology]

Conventional distance measurement methods for measuring the distance to an object include methods that emit ultrasonic waves or laser light and capture the reflected waves from the object, and use this to measure the distance. According to these methods, the absolute distance to an object can be measured without contact.

[Problem that the invention seeks to solve]

However, distance measuring methods using ultrasonic waves or laser light have the following problems.

(1) What can be measured using these methods is basically the distance from the observation point to a point on the object onto which the ultrasonic wave or laser beam is projected. Therefore, if these methods are used to measure the distance between the vehicle and the preceding vehicle while the vehicle is running, it is difficult to always measure the distance reliably due to oscillations and changes in relative position caused by the drive.

(2) When trying to measure the distance to a distant object, it is necessary to increase the output of ultrasonic waves or laser light, which causes cost and space problems.

(3) In the case of ultrasonic waves, the beam cannot be focused due to directivity, and therefore long-distance measurements are difficult.

Although laser light can be highly directional, another problem is that the beam can be dangerous if it enters the human eye.

[Means for solving problems]

In order to solve the problems of the prior art, the present invention has the following points:
The inter-vehicle distance measurement method for detecting changes in the relative distance to a moving object includes the following means.

The apparatus includes means for performing spatial differential processing on an image of the object to detect a horizontal component and then binarizing the detected horizontal component.

This provides means for determining the significant width in the vertical direction in the binarized image.

A means for determining a significant width in the horizontal direction in the vertical significant region thus determined is provided.

Furthermore, the amount of change over time in the width of this horizontal direction significant area is detected, the direction of increase or decrease in this time change is detected to determine whether the object is approaching or moving away, and/or the amount of change over time is increased. It includes means for detecting that the direction exceeds a predetermined value and determining abnormal approach.

[For production]

The object is imaged and the image data is processed to determine the width of the object on the screen (apparently the width of the vehicle). In this case, the vehicle width on the screen is determined by extracting features in the horizontal direction through image processing, taking advantage of the fact that the vehicle's width has a horizontal edge component. Such measurements are performed in real time, and the direction of increase or decrease in the temporal change in vehicle width on the screen is determined, and a display indicating whether the preceding vehicle is moving away or approaching is performed. Furthermore, when it is detected that the amount of change exceeds a predetermined value in the increasing direction, an alarm or display indicating abnormal approach is issued.

[Embodiment of the Invention] FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, an imaging device 1 is for imaging an object, particularly a preceding vehicle, and is, for example, a conventional ITV camera using a camera tube, or a CO2 camera using a CCD element.
A D camera etc. can be used. The imaging device 1 is installed at an appropriate location in the vehicle 2 at a position that does not interfere with the driving of the vehicle and where a sufficient forward field of view can be secured. A/
The D converter 3 performs analog-to-digital (A/D) conversion on the video signal from the imaging device 1, and inputs the converted signal to the processing device 4. The processing device 4 has a memory 5 in which the digitized video signal from the A/D converter 3 is stored in a memory 5. The processing device 4 further includes a computer 6, which calculates the vehicle width of the preceding vehicle from the data stored in the memory 5 at every predetermined time (sampling time), compares it with the vehicle width measured last time, and determines the vehicle width. Find the direction and speed of change, and determine if the vehicle width increases (+
) direction or a decreasing (-) direction, and an abnormal approach warning signal indicating that a predetermined threshold value has been exceeded in the increasing direction. Based on the calculation results of the computer 6, the inter-vehicle distance meter 7 displays a display indicating that the preceding vehicle is approaching in response to a secondary direction signal, and indicates that the preceding vehicle is moving away in response to a unidirectional signal. When an abnormal approach warning signal is generated,
A warning is issued by displaying the message ``Slow down!'' and sounding a buzzer.

FIG. 2 is a diagram illustrating a processing algorithm of the computer 6 in the processing device 4. As shown in FIG. In the method of the present invention,
This technology utilizes the fact that the rearview image of the vehicle has horizontal edge components when traveling straight to speed up processing, and the conventional technology measures distance by observing a point on the body of the vehicle in front. Unlike the previous method, this method is characterized in that reliable processing results can be obtained because the preceding vehicle is understood and processed as a certain area on the screen.

Figure 2 (al indicates image data, and for ease of understanding, only the vehicle 11 is shown. Image processing technology is applied to this image data to create the vehicle 1 as shown in Figure 2 (b).・Extract only l and set a predetermined sampling time interval Δt
The vertical direction is A-B, and the horizontal direction is C.
−D) to calculate the inter-vehicle distance. Note that the image data here is assumed to be a grayscale (black and white) image,
It is assumed that each pixel has brightness information.

FIG. 2(C) is a flowchart showing the flow of processing. Hereinafter, the processing will be explained step by step based on these figures. FIGS. 3(a) to 3(el) show the processing results at each stage.

■, Detection of horizontal component (step Sl) Fig. 2 (a)
A differential image is obtained by performing a spatial differentiation process on the image data to obtain a change in brightness in the vertical direction. This image represents the horizontal component of the vehicle body image.

■, Binarization (Step S2) Differential pixels (hereinafter simply referred to as pixels) of each point in the image obtained with Ste-nobu S1 are converted into "
For example, when each pixel value is xl; each pixel value is set to 0 (white).

When each pixel value ≧l; Let each pixel value be 1 (black).

FIG. 3 (al is an example of an image obtained by binarizing in this manner after detecting the horizontal component.

■Determination of A, B (Step S3) For the image obtained in Fig. 3(a), the value (0 or 1) of each pixel is integrated in the horizontal direction, and the result is as shown in Fig. 3(b). Create a horizontal histogram. For this, a certain threshold value m is determined, the image is scanned vertically from the top and bottom edges, the histogram is compared with the threshold value m, and the first points that satisfy the histogram value ≧m are designated A and B, respectively. . When A and B are determined in this way, a processing area is determined with A and B as upper and lower limits, as shown in FIG. 3(C).

■, Determination of C, D (Step S4) Figure 3 (In the processing area shown in C1, the value (O or 1) of each pixel is integrated in the vertical direction. Figure 3 (d)
Create a vertical histogram as shown. Next, find the average value of the histogram in the vertical direction, find the widest range that satisfies "histogram value ≧ average value", and set the points at both ends as C and D, respectively. Since C and D are determined by calculating the average value in this way, even if there is noise as shown by a in the vertical histogram, it will not be detected.

Figure 3(e) is sized by A, B and C, D determined in this way. This shows an extracted image of a vehicle. In these FIGS. 3, the distance between C and D indicates the apparent vehicle width W on the screen.

(2) Calculation of relative inter-vehicle distance (step S5) Letting the vehicle width of the preceding vehicle at time t be w (t), it can be expressed as W (t)=DC.

Furthermore, from now on, the sampling time Δ is changed to the previous vehicle width W(
When W (t)-W (-Δt)<Q, it is determined that the preceding vehicle is moving away.

When W (t)-W (-Δt)>Q, it is determined that the preceding vehicle is approaching.

C. Set a certain threshold value n (n>O) and W
When (t)-W (-Δt)>n, it is determined that there is an abnormal approach state.

Depending on the results of judgments A to C, as mentioned above, the distance meter displays whether the vehicle in front is approaching or receding, and also sounds a buzzer to warn of abnormal approach.
It is possible to display a message saying "slow down".

FIG. 4 shows an example of image processing according to the present invention, and (a) shows original image data.

However, this example emphasizes the outline of the image data, and is different from actual image data. Figure 4 (mountain) shows the image that has been binarized after detecting the horizontal component;
This corresponds to the image data that has been processed in Figure fa). Figure 4(C) shows the extracted image of the vehicle, and Figure 3(C) shows the extracted image of the vehicle.
This corresponds to the image data after the processing in 61 is completed.

As seen in FIG. 4, image data of a vehicle in a natural environment is clearly extracted, demonstrating the effectiveness of the present invention.

Further, as an application example of the present invention, when the width of the preceding vehicle is known, it is also possible to calculate the absolute inter-vehicle distance as follows.

It is assumed that the imaging device (2) shown in FIG. 1 is located at a height H above the ground and is mounted with the optical axis of the imaging lens parallel to the ground. As shown in Fig. 5, the center of the lens with focal length f is taken as the origin of the three-dimensional coordinate system, and a point P (x, y,
z) is centrally projected onto the image plane by p (
φ+n), the following relationship holds true.

φ=f − n=f − The coordinates on the image plane corresponding to points C and D in Fig. 3 are respectively (φI+nl) + (φZrnz), and the corresponding coordinates on the three-dimensional space are ,,Zl
), (Xz, Yz, Zi, assuming that the road surface is flat and a straight road, the following relationship holds true.

nl=n! Y,=YZ Furthermore, the following relationship exists between the vehicle width <xz -XI) and the inter-vehicle distance Z.

Xt - XI = (φ2 - φ1) In this case, since the vehicle width (Xz XI) is known, the inter-vehicle distance Z is determined as follows.

φ2−φi [Effects of the Invention] As explained above, according to the present invention, the relative distance to an object (inter-vehicle distance) can be measured in real time using only one imaging device. . In this case, taking advantage of the fact that the vehicle's rear view has a horizontal edge component when traveling straight,
Processing speed can be increased. Unlike conventional techniques that measure the distance between a point on the preceding vehicle and an observation point, the method of the present invention grasps the vehicle as an image of a certain area on the screen and processes it, resulting in highly reliable measurement results. can be obtained. As an application of the present invention, it is also possible to calculate the absolute inter-vehicle distance if the width of the preceding vehicle is known.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an embodiment of the present invention, and FIG. 2 (a) to (
C1 is a diagram showing the algorithm of convenience store processing in the processing device, FIGS. 3(a) to (el) are diagrams explaining each stage of processing in the present invention, and FIGS. Fig. 5 is a diagram showing an example of image processing according to the present invention, and Fig. 5 is a diagram showing a coordinate system when measuring an absolute inter-vehicle distance.
...Vehicle, 3...A/D converter, 4...
...Processing device equipped with each functional means, 7...
Inter-vehicle distance meter, 11... Vehicle. Patent Applicant: Suzuki Motor Co., Ltd. Figure 1 Figure 2 Figure 3 (a) (b) = -A (e) 222D Figure 4 (C) (Medium car's f-nail drawing) I -7°γ-;;2Figure 5

Claims (1)

[Claims] (1) In an inter-vehicle distance measurement method that detects a change in relative distance to a moving object, means for performing spatial differential processing on an image of the object to detect a horizontal component and then binarizing it; means for determining the significant width in the vertical direction in the valued image; means for determining the significant width in the horizontal direction in the significant area in the vertical direction; and the amount of change in the width of the significant horizontal area over time; The direction of increase or decrease in change is detected to judge whether the object approaches or moves away from the object, and if the amount of change over time exceeds a predetermined value in the increasing direction, this is detected to determine abnormal approach. An inter-vehicle distance detection method characterized by comprising means.