JPH0444200A - Illegal parking confirming device - Google Patents

Abstract

PURPOSE:To surely detect the existence of illegal parking vehicles which are parked with lights extinguished at night by detecting movement of a vehicle by movement of a bright point of the head light or the like and deciding the rear-end collision avoiding operation of the succeeding vehicle to settle illegal parking vehicles. CONSTITUTION:A photographing means 1 has the exposure adjusted so that only high-luminance light of the head light or the like is extracted without halation, and the means 1 photographs the detection object section of illegal parking vehicles set on a road. A picture signal of the prescribed section on the road photographed by the photographing means 1 is converted to digital density data by an AD converter 2. A vehicle tracing part 3 traces movement of a vehicle by movement of the bright point of the head light or the like, and an avoiding operation decision processing part 4 decides the rear-end collision avoiding operation of the succeeding vehicle based on movement information of the bright point detected by the vehicle tracing part 3, thus detecting and confirming the existence of illegal parking vehicles.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an illegal parking confirmation device that detects the presence of an illegally parked vehicle at night based on a rear-end collision avoidance operation of a following vehicle.

[Conventional technology]

Conventionally, as a method of detecting a parked vehicle using image processing, there is a method of determining the presence or absence of a parked vehicle based on changes in brightness due to the vehicle in an image taken of a road surface, as shown in Japanese Patent Application Laid-open No. 174198/1983, for example. Are known.

In addition, as a method of measuring vehicle traffic flow, the road to be detected is photographed using an average photometry method (a method that determines the appropriate aperture value based on the average illuminance within the screen) using an ITV camera.
A method of detecting a vehicle has been put into practical use by subjecting the video signal to processing such as binarization and then extracting the shape characteristics of the vehicle using a predetermined threshold value.

[Problem to be solved by the invention]

However, the former method, which detects parked vehicles based on changes in the degree of detail in images taken of the road surface, is practical during the day when it is easy to distinguish between the background and the vehicle, but at night, most roads are illuminated by road lights. Because the background is dark, the vehicle is absorbed into the background, making it difficult to distinguish between the background and the vehicle.

On the other hand, in the latter method, which uses a threshold value to extract the vehicle's shape characteristics from video signals shot using the average photometry method, since there is a large difference in brightness between daytime and nighttime, it is necessary to shoot at night in the same way as during the daytime. I use a high-sensitivity ITV camera, and have to close down the lens aperture during the day and open the lens aperture at night.

However, at night, a vehicle's HeraDry 1- has a fairly high brightness, so if you photograph it with a high-sensitivity camera and the lens aperture is open, the brightness will be saturated on the imaging surface. (hereinafter referred to as "halation")
Therefore, even if processing such as binarization is performed on this image, the halation region will occupy most of the image, causing the problem that the shape characteristics of the vehicle that should be detected will disappear.

Furthermore, the higher the sensitivity of the imaging surface, the more the effects of halation will extend to surrounding vehicles. For example, when the light from the headlights hits another vehicle and enters the imaging surface of the ITV camera, it is possible to distinguish whether the light is direct or reflected from the headlights from the reduction in brightness caused by reflection. When the sensitivity of the imaging surface is high as described above, there is a problem in that this reflected light portion also becomes a halation state, making it impossible to extract the geometrical features of the vehicle that is hit by the reflected light.

The present invention has been made in view of the above problems, and its purpose is to provide an illegal parking confirmation device that can reliably detect illegally parked vehicles at night.

[Means to solve the problem]

In order to achieve the above object, the present invention detects the presence of illegally parked vehicles parked with their lights turned off at night based on the movement of high brightness points such as the headlights of the following vehicles. As shown in Fig. 1, its basic configuration includes a photographing means 1 whose exposure is adjusted to extract only high-intensity light such as headlights, and an image signal of a predetermined area on the road photographed by the photographing means 1. An AD converter 2 that converts into density data, a vehicle tracking unit 3 that detects the movement of bright spots such as headlights from the digital density data of the AD converter 2, and a vehicle tracking unit 3 that detects the movement of bright spots such as headlights from the digital density data of the AD converter 2. It is comprised of an avoidance operation determination processing section 4 that determines rear-end collision avoidance operation of a following vehicle based on movement information and outputs an illegal parking confirmation signal.

[For production]

The detection principle of the device of the present invention will be explained with reference to FIG.

The photographing means 1 is exposure-adjusted to extract only high-intensity light such as headlights so as not to cause halation, and photographs a detection target area of illegally parked vehicles set on the road.

FIG. 2(a) shows an image f (T) taken by this photographing means 1 at time T during nighttime. In the figure, the oval-shaped portion is the headlight of the vehicle. Also, the same figure (b
) and (C) are images f(T+1) and f(T+2) at times T+1 and T+2, respectively. Note that the times T, T+1, and T+2 have a relationship of T<T+1<T+2, and each time interval is approximately several tens of milliseconds to several hundred milliseconds.

Now, Figure (a) and Figure (
5(t)-f (t)-f (t+1) for the image of c)
(1) Assume that the following calculation is performed. Here,
f(t) is the digital image at time t that has been AD converted by AD converter 2, and f (t+1) is the digital image at time t −1
The digital image s (t) at −1 is a difference image between the two digital images f (t) and f (t+1).

The results of the difference calculation of equation (1) using the images in Figure 2 (a) and (t)) are shown in Figure 2 (d).
The change in density value in the vertical direction of the difference image in d) is shown in FIG. 3(f).

When shooting at night with exposure adjusted so that light from headlights does not cause halation, the brightness of the outside world becomes almost pitch black. That is, in the digital density value quantized by the AD converter 2, the brightness in the outside world is almost 0 level. Therefore, now this figure 2 (a)
When the vehicle at the P0 position moves to the P2 position in Figure (b), the vertical density value change in the difference image 5(t) (Figure (d)) is shown in Figure (f). Like the original P5
A positive density value component (+A) is generated at the position, and a negative density value component (-8) is generated at the moved position P2.

Furthermore, using the images in Figures 2(b) and (C), (1
) The result of performing the difference calculation of the equation (2) is shown in Figure (e), and the change in density value in the vertical direction of the difference image in Figure (e) is shown in Figure (6). As is clear from the figure, a positive density value component (+A) is generated at the original P2 position, and a negative density value component (-A) is generated at the moved P3 position.

As is clear from comparing Figure (f) and Figure (g), Figure (
A negative concentration value component of (-A) occurs at the P2 position in Fig. That is, the image data obtained by converting the image signal photographed by the photographing means 1 of the present invention into digitized data by the AD converter 2 is sampled at time intervals of approximately several tens of milliseconds to several hundreds of milliseconds, and If the forward difference processing given by equation (1) is performed continuously by subtracting the next image from the previous image, if the vehicle is moving continuously, It is possible to detect that the vehicle is moving in the direction where the negative density value component exists, starting from the positive density value component.

Furthermore, when the same vehicle is moving, the position of the positive density value component in the difference image of interest is always the same as the position of the negative density value component in the previous difference image. Therefore, by incorporating this into the conditions for vehicle determination, it is possible to track the same vehicle.

Note that the difference calculation in equation (1) described above is a forward difference calculation in which the next image is subtracted from the previous image.
Conversely, it is of course possible to perform a backward difference calculation in which the previous image is subtracted from the next image. In this case, the sign of the direction of movement is simply reversed.

The vehicle tracking unit 3 tracks the movement of the vehicle from the movement of bright spots such as headlights as described above.

Then, based on the movement information of the bright spot detected by the vehicle tracking section 3, the avoidance operation determination processing section 4 determines the rear-end collision avoidance operation of the following vehicle, and detects and confirms the presence of an illegally parked vehicle.

Generally, when a vehicle is illegally parked on one side of the road, blocking part of the lane, the following vehicle must turn the steering wheel in front of the parked vehicle to avoid rear-ending the parked vehicle. or if you are too close to a parked vehicle to operate the steering wheel in time,
- After stopping in front of a parked vehicle, it is normal to perform actions to avoid a rear-end collision, such as backing up the vehicle. Therefore, if such a rear-end collision avoidance operation is detected, it is possible to detect that an illegally parked vehicle is present in front of the following vehicle that has performed the rear-end collision avoidance operation. Various methods can be used to determine this rear-end collision avoidance operation, some of which will be exemplified below.

That is, in general, when the steering wheel is turned to avoid a rear-end collision, the direction of travel of the vehicle typically tilts significantly from the direction along the road to the left or right. Therefore, focusing on this point, for example, a reference vector for determining the direction of travel of a vehicle is set parallel to the road, and the inclination of this reference vector and the brightness of headlights that indicate the direction of travel of the following vehicle are used. The magnitude of the inclination of the moving vector of the point is compared, and when the inclination of the moving handle of the bright point becomes larger than the inclination of the reference vector, it can be determined that the following vehicle has performed a rear-end collision avoidance operation.

In addition, in consideration of the fact that when the steering wheel is turned to avoid a rear-end collision, the vehicle often veers into the oncoming lane or adjacent lane and takes a detour, for example, if the vehicle veers out along the center line or lane marking line of the road, A reference line is set for determination, and the position coordinates of this reference line are compared with the position coordinates of a bright spot such as a headlight that indicates the location of the following vehicle. When the vehicle protrudes beyond the position coordinates, it may be determined that the following vehicle has performed a rear-end collision avoidance operation.

Additionally, if the vehicle is so close to a parked vehicle that it is not possible to operate the steering wheel in time, the following vehicle may stop in front of the parked vehicle and then perform maneuvers to avoid a rear-end collision, such as backing up the vehicle. In view of this, it is sufficient to monitor the polarity reversal of the movement vector of the bright spot indicating the direction of travel of the following vehicle, and determine that the following vehicle has performed a rear-end collision avoidance operation when the polarity of the movement vector is reversed.

Furthermore, when the steering wheel is turned in order to avoid a rear-end collision, the following vehicle usually turns the steering wheel significantly to the outside of the parked vehicle and then turns the steering wheel sharply again to return to the original driving lane. Therefore, if we focus on this point, when it is detected that the lateral component of the movement vector of a bright point such as a headlight that indicates the direction of travel of the following vehicle has changed to the component in the direction of returning to the original lane, the following vehicle will collide with the driver. It can also be determined that an avoidance action has been taken.

Furthermore, various determination methods can be employed, such as using some of the above-described determination processes in combination.

〔Example〕

Examples of the present invention will be described below.

FIG. 3 shows a block circuit diagram of one embodiment of the illegal parking confirmation device of the present invention.

Photographing means 1 includes an ITV camera 101 and a variable aperture lens 1
02. Consisting of an aperture control circuit 103, the aperture control circuit 103 detects the peak value of the luminance signal in the image signal output from the ITV camera 101, and performs negative feedback control to control the variable aperture lens 102. This allows you to photograph only high-brightness lights such as headlights. Note that the ITV camera 101 is fixedly arranged on a support column as illustrated in FIG. 4 in order to photograph the same position on the road surface.

The image signal output from the ITV camera 101 is passed through a low-pass filter 5 and then sent to the AD converter 2. The low-pass filter 5 is inserted to remove noise components in the image signal captured by the ITV camera 101 and to prevent aliasing noise from occurring in the AD converter 2, which will be described later, and is designed to satisfy the sampling theorem. Its cutoff frequency is set to .

The AD converter 2 samples the image signal from which noise components have been removed by the low-pass filter 5, and converts it into multi-gradation digital density data.

Using the digital density data obtained as described above, the processes (1) to (8) described below are repeatedly executed at fixed time intervals in the vehicle tracking section 3 and the avoidance motion determination processing section 4. The presence of an illegally parked vehicle is detected from the rear-end collision avoidance actions of the following vehicle.

(1) In the initial state, there is no image data in any of the image memories 301 and 302 in the vehicle tracking unit 3, so the memory control unit 6 outputs the image memory selection signal d to the image memory 301, Output image data a from the AD converter 2 is written into one image memory 301 through an image memory input path b.

(2) After this writing, the memory control unit 6 immediately outputs the timer start signal f and starts the sampling timer 7 that defines the sampling interval t set to several tens of milliseconds to several hundred milliseconds.

(3) After this time interval t has elapsed, the sampling timer 7 outputs a time-up signal e to the memory control unit 6.

(4) Upon receiving this time-up signal e, the memory control unit 6 outputs an image memory selection signal C to the image memory 302, and after a certain time interval elapses,
Output image data a from the D converter 2 is written into the other image memory 302 via an image memory input bus b.

(5) After writing to this image memory 302 is completed,
The memory control section 6 outputs a differential processing signal i to the differential processing section 303. The difference processing unit 303 that has received this difference processing signal i reads the contents of the image memories 301 and 302 from the image memory output bus g+h, performs the difference calculation according to the above-mentioned equation (1), and sends the result of the calculation to the output bus j.
The data is stored in the output memory 304 through. After the difference calculation between the image data of the image memories 301 and 302 is completed, a difference processing end signal is sent to the microprocessor 401 forming the avoidance motion determination processing section 4! Output.

(6) The microprocessor 401 receives this differential processing end signal! When this happens, the process shown in the flowchart of FIG. 5, which will be described later, is executed to detect the presence of a parked vehicle based on the rear-end collision avoidance operation of the following vehicle. After this processing is completed, a processing end signal m is output to the memory control unit 6.

(7) Upon receiving this processing end signal m, the memory control unit 6 transfers and writes the contents of the image memory 302 to the image memory 301 via the image memory bus b, and further sends a timer start signal f to the sampling timer 7. , start the timer.

(8) Repeating each process of (3) to (7) above.

FIG. 5 is a flowchart of the processing of the microprocessor 401 after the differential processing end signal l described in the above processing (6) is input.

The operation of the microprocessor 401 will be described below with reference to FIG.

In the moving direction determination process of step [1], the ITV
By scanning the difference image obtained by the difference processing unit 303 for the detection area on the road taken by the camera 101, a polarity reversal component in which the density value is reversed from positive to negative and from negative to positive is searched. , the position of the polarity-inverted component is detected. Furthermore, check whether the position of the positive component (P2 in Figure 2 (to)) is the same as the position of the negative component (P2 in Figure 2) of the previous difference image. Determine the movement start point P2 of the vehicle by determining,
Next, by determining whether the position of the negative component (P3 in FIG. 2(f)) is the same as the position of the positive component of the next difference image (not shown), the end point P3 of the vehicle's movement is determined. is determined, the moving direction from the movement start point P2 to the movement end point P3, that is, the movement vector of the vehicle is determined, and the slope of the movement vector is calculated.

This movement vector can be determined from the position coordinates before and after the movement, and its slope can be determined from the ratio of the difference between the coordinate values before and after the movement. To explain this using Figure 6 as an example, if the following vehicle performs a rear-end collision avoidance maneuver as shown in the solid line (trajectory of the right headlight) in the diagram to avoid a rear-end collision with an illegally parked vehicle in front, the movement start point Let the position coordinates of (XI,
Yl ), the position coordinates of the movement end point are (XZ , Yz
), the movement vector at this time is also, and the slope of the movement vector at this time is Y2-, y+ l/l XZ -XI +
It is determined by (3).

In step [2], it is determined whether or not the following vehicle is moving based on the movement vector obtained as described above. If it is determined that the following vehicle is moving, step [3] is carried out to determine if the vehicle is moving. ” to remember.

In the next step [4], a comparison is made between the slope of the moving Peltle obtained above, lY2-Yl l/lX2-XIl, and the slope of the reference vector, 1ΔYs/ΔX81. If the inclination is greater than the inclination to the reference vector, it is determined that a rear-end collision avoidance operation is present, that is, the following vehicle has significantly changed its direction of travel in order to avoid a rear-end collision with the illegally parked vehicle in front, and step [
5], an illegal parking confirmation signal indicating the existence of an illegally parked vehicle is output.

Incidentally, the inclination 1ΔYs /ΔXs l of the reference vector for determining the presence or absence of rear-end collision avoidance operation is a reference vector set almost parallel to the road converging toward the rear of the screen, as illustrated in FIG. is given by the slope of . This reference vector can be set in advance on the screen once the road to be detected and its detection area are known.
The slope lΔYs/Δxs 1 is also uniquely determined. Therefore, this value of 1ΔYs/ΔXs 1 may be stored in advance in the microprocessor 401 as a comparison standard.

If it is determined in step [2] of FIG. 5 that there is no movement of the vehicle, it is determined in step [7] whether or not there was movement of the vehicle during the previous process.

If it was "moved" during the previous process, it is determined that the vehicle is parked at the relevant position during the current process, and after storing the parking position coordinates in step [8],
In step [9], an illegal parking detection signal is output.

In addition, in step [7], the movement "
If it is "None", it is determined that the vehicle has been parked at that position before, and the vehicle jumps to step [6].

When the above processing is completed, the microprocessor 401 outputs a processing end signal m to the memory control unit 6 in step [6], and returns to the above-mentioned processing (7). In this way, the microprocessor 401 repeatedly executes the process shown in FIG. 5 every time it receives the difference processing end signal l from the difference processing unit 303, and detects illegally parked vehicles.

In the case of the embodiment shown in FIG.
Since steps [7] to [9]) are processed in combination, illegally parked vehicles can be detected with higher reliability.

In the above embodiment, the determination process of the rear-end collision avoidance operation of the following vehicle in step [4] is performed based on the inclination of the movement vector of the following vehicle l Yz Y+ l / l XzXI
l and the inclination to the preset reference handle ΔYs /Δ
Although this was done by comparing the size with xs 1, it is also possible to adopt a determination method using a reference line L as shown in FIG. 8 instead.

That is, in general, when the steering wheel of a following vehicle is turned to avoid a rear-end collision, the vehicle often goes off the center line and takes a detour as shown in FIG. Therefore, a reference line L for determining whether the vehicle is protruding is set along this center line, and the position coordinate of this reference line L is
s and the position coordinate x2 of a bright spot that indicates the location of the following vehicle, and when X z < , Illegal parking confirmation signal (step [51
) should be output.

Furthermore, if the vehicle is so close to the parked vehicle that it is too late to operate the steering wheel in time, the following vehicle may temporarily stop in front of the parked vehicle and then back up. Therefore, as a determination process for a rear-end collision avoidance operation, the existence of illegal parking can also be determined by detecting that the polarity of the movement vector of the bright spot is reversed from positive (forward movement) to negative (backward movement).

Furthermore, as is clear from the 6th episode, when the steering wheel is turned to avoid a rear-end collision, the following vehicle turns sharply to the outside of the parked vehicle and then sharply turns the steering wheel again at the (x3.y3) position. It is normal to return to the original driving lane. Therefore, at this turning point N(X3, Y3), it is detected that the lateral component (X4-X3) of the movement vector has changed to a positive value (X4-X3)>0 indicating returning to the original lane. At times, it may be determined that the following vehicle has performed a rear-end collision avoidance operation.

FIG. 9 shows another embodiment of the processing of the microprocessor 401. In the figure, the same step numbers as in FIG. 5 indicate the same processing.

The process shown in FIG. 9 differs from the process shown in FIG. 5 in that a process for determining the frequency of rear-end collision avoidance operations in step [4'] is added.

Generally, when a rear-end collision avoidance operation is determined at short intervals of several tens of milliseconds, the determination process is executed multiple times until the rear-end collision avoidance operation of the following vehicle is completed. Therefore, a predetermined judgment value N (N is an arbitrary integer value 1, 2, 3, etc. according to the system specifications) is set in advance,
When the number of times the rear-end collision avoidance operation is determined in step [4] exceeds this determination value N, an illegal parking confirmation signal is output. By doing so, it is possible to more reliably identify illegally parked vehicles.

In the above embodiment, the movement vector is determined from the trajectory of the right headlight of the vehicle, but the present invention is not limited to this. Various methods can be adopted, such as finding it from the locus of the center of.

In addition, an automatic adjustment method using negative feedback control was adopted to adjust the exposure of the ITV camera 101 as a photographing means.
When the external illuminance is detected and its value falls below the set value,
It is also possible to automatically change the aperture to a pre-set lens aperture. Furthermore, various methods can be employed, such as adjusting exposure by remote control.

In the embodiment described above, the positive direction of the x-y coordinate axis is
Although the case where the coordinates are taken in the directions shown in FIGS. 8 to 8 has been described as an example, the positive and negative directions of the coordinate axes can be taken freely. It is clear that the same method can be achieved by changing the sign of each value and the direction of the inequality sign in the above-mentioned rear-end collision avoidance operation determination process depending on how the coordinate axes are taken.

〔Effect of the invention〕

As is clear from the above description, when using the illegal parking confirmation device of the present invention, the movement of a vehicle at night is detected from the movement of a bright spot such as a headlight, and based on the movement information of the bright spot, the following Since the vehicle is now determined to be an illegally parked vehicle by determining the vehicle's rear-end collision avoidance behavior, it is now possible to reliably detect the presence of illegally parked vehicles, such as those parked with their lights off at night, which was previously difficult to do. .

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the basic configuration of the illegal parking confirmation device of the present invention, FIG. 2 is a diagram showing the vehicle detection principle of the present invention, and FIG. 3 is a block circuit diagram of one embodiment of the present invention. 4 is a diagram showing an example of the installation of the TTV camera in the above embodiment, FIG. 5 is a flowchart of the processing of the microprocessor in the above embodiment, FIG. FIG. 7 is an explanatory diagram of the reference vector, FIG. 8 is an explanatory diagram of the reference line, and FIG. 9 is a flow chart showing the other side of the microprocessor processing. DESCRIPTION OF SYMBOLS 1... Photographing means, 2... AD converter, 3... Vehicle tracking section, 4... Avoidance motion determination processing section, K... Reference vector, L... Reference line.

Claims (3)

[Claims] (1) A photographing means whose exposure is adjusted to extract only high-intensity light such as headlights, and an AD converter which converts the image signal of a predetermined area on the road photographed by the photographing means into digital density data; a vehicle tracking unit that detects movement of a bright spot such as a headlight from digital density data of the AD converter; and a vehicle tracking unit that determines rear-end collision avoidance behavior of a following vehicle based on movement information of the bright spot detected by the vehicle tracking unit. 1. An illegal parking confirmation device comprising: an avoidance operation determination processing section that outputs an illegal parking confirmation signal. (2) In the illegal parking confirmation device according to claim (1),
An apparatus for determining illegal parking, wherein the avoidance operation determination processing unit determines the rear-end collision avoidance operation of a following vehicle by comparing the inclination of the moving vector of the bright spot with the inclination of a reference vector set on the road. (3) In the illegal parking confirmation device according to claim (1),
The illegal parking determination device is characterized in that the avoidance operation determination processing unit determines the rear-end collision avoidance operation of the following vehicle by comparing the position coordinates of the bright spot with the position coordinates of a reference line set on the road.