JP2953176B2 - Vehicle detection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle detecting apparatus for measuring the position, number, speed and the like of vehicles traveling on a road by image processing, which is used in a road traffic system.

[0002]

2. Description of the Related Art Conventionally, radar-type and ultrasonic-type vehicle sensors have been used as vehicle detection devices. In recent years, vehicle detection devices to which image processing technology is applied have been developed and operated, but all of them are for measuring the number of vehicles and the vehicle speed.
An optical vehicle detection device is used for detecting the position of a vehicle.

[0003]

However, among the above conventional examples, the radar type, ultrasonic type vehicle detector and image processing type vehicle detection device have a problem that the position detection accuracy for a high-speed vehicle is difficult. . On the other hand, the optical vehicle detection device has a problem that the number measurement accuracy is difficult.

The present invention has been made to solve the above-mentioned conventional problems, and can obtain high vehicle position detection accuracy, vehicle speed measurement accuracy, and vehicle number measurement accuracy. It is another object of the present invention to provide an image processing type vehicle detection device capable of improving economy.

[0005]

According to the present invention, there is provided a synchronous signal generating means for supplying a synchronous signal.
And the number of effective scanning lines can be arbitrarily changed by the synchronization signal.
Levi camera device and vehicle image captured by TV camera
Storage means for storing data, and currently captured image data
Processing between the image data and the previous image data stored in the storage means
And a subtraction processing means for outputting a pixel corresponding to the difference
And this difference pixel in the horizontal and vertical directions of the screen, respectively.
Profile to obtain profile data by adding
Processing means, and a vehicle width component from the profile data.
By detecting the leading edge of the vehicle, the position and speed of the vehicle
Calculation means for measuring the

[0006]

Therefore, according to the present invention, a synchronizing signal generator is provided.
Control the number of effective scanning lines of a TV camera device
This can reduce errors in vehicle position detection.
In addition, the subtraction of the current image data and the previous image data
Performs the processing and determines the pixels corresponding to the difference from the horizontal
Profile data obtained by adding each in the vertical direction
Detects the vehicle width component and the vehicle front edge from the
Position and speed can be calculated .

[0007]

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

FIG. 1 is a schematic block diagram showing a vehicle detecting device according to an embodiment of the present invention. In FIG. 1, 11
Is a television camera device that captures an image of a traveling vehicle and sends out a video signal, 12 is a synchronization signal generator that supplies a synchronization signal to the television camera device 11, and 13 generates a voltage for controlling the lens of the television camera device 11. A lens control voltage generator 14 is an AD converter for AD converting the video signal transmitted from the television camera device 11, 15 is a frame memory for storing the conversion result of the AD converter 14, that is, image data, and 16 is a frame difference. This circuit performs a difference process between the image data that has been A / D converted and the image data that has been A / D converted one frame before and stored in the frame memory 15. Reference numeral 17 denotes a profile circuit, which executes a so-called profile process of counting the number of pixels determined to be significant by the frame difference, that is, the number of pixels having a difference in the vertical and horizontal directions. 18 is a profile circuit 1
7, a profile memory for storing profile values, and 19 is a microprocessor (hereinafter, referred to as a profile memory).
The CPU determines the vehicle from the profile data stored in the profile memory 18, calculates the amount of measurement, processes the data in the frame memory 15, and controls the lens of the television camera device 11.

The above configuration will be described in further detail below together with its operation. The television camera device 11 operates in synchronization with the synchronization signal generated by the synchronization signal generation unit 12, and can operate with an arbitrary number of effective scanning lines. The number of effective scanning lines is N, the number of scanning lines in the retrace period is N _B ,
Assuming that the horizontal scanning time is H, the time T required to capture one screen is represented by the following equation.

[0010] T = and H (N + N _B), when the vehicle speed is V, the moving distance L of the vehicle between one screen imaging is expressed by the following equation.

L = VT = VH (N + N _B ) Since N _B = 19 and H = 63.55 μsec, N = 6
4. If V = 120 km / h, L = 17.6 cm. That is, according to this method, a high accuracy of a vehicle position error in one-screen imaging at a maximum of 17.6 cm at 120 km / h can be obtained.

The image taken in such a manner that the error of the vehicle position due to the vehicle speed is kept small is converted by the AD converter 14 into 8
The data is converted into digital data of 256 gradations, stored in the frame memory 15 and sent to the frame difference circuit 16. The frame difference circuit 16 performs a subtraction process between the transmitted current frame data and the frame data of one frame before stored in the frame memory 15 in advance, and determines a pixel having a significant difference in the current data. Extract. Also, the current data is stored as frame data as needed. The significant pixels thus extracted are added by the profile circuit 17 in the horizontal and vertical directions of the screen, and the number of significant pixels is calculated. When a vehicle is present, a characteristic signal of the vehicle is detected as profile data, and the profile data is stored in the profile memory 18.

FIG. 2 schematically shows data in the profile memory 18 for detecting the vehicle 21. The sum of the difference pixels in the horizontal direction is a frame difference horizontal profile 22, and the sum of the difference pixels in the vertical direction is a frame difference vertical profile 23. Frame difference circuit 16
Is a subtraction process between the currently captured image data and the image data captured one frame before, so that a portion of the moving object having a large change in luminance is extracted. That is, the horizontal line component of the vehicle 21 is detected as a horizontal profile value, and the vertical line component is detected as a vertical profile value. Also, dark areas such as shadows and windshields are detected as black components, and bright areas such as bumpers are detected as white components.

The CPU 19 processes the profile data stored in the profile memory 18 and measures the required amount of the vehicle 21 such as the position, speed, and number of vehicles. That is, when the vehicle 21 is present, both ends of the data of the frame difference vertical profile 23 are searched to determine the length equal to or longer than the vehicle width (the shadow often extends horizontally and becomes longer than the vehicle width). Can be measured. In FIG. 2, the L point of the black difference is detected as the left end, and the R point of the white difference is detected as the right end. Depending on how the shadow appears, the R point may become a black difference or both may become a white difference. In any case, when the following relationship is satisfied, a moving object having a size corresponding to the vehicle 21 exists. Can be determined to be.

W = RL−vehicle width Next, the CPU 19 searches the data of the frame difference horizontal profile 22 to detect the front edge of the vehicle 21. As shown in FIG. 2, when data is searched from the lower side of the drawing in the direction of the arrow, the feature point A of the black difference is detected first. Here, when the difference detected first is a black difference, the CPU unit 19 determines that the difference is a shadow and continues the search to detect a white difference B. This is the front edge of the vehicle 21. A similar vehicle front edge characteristic signal is detected in the black difference C and the white difference D of the horizontal profile data.
When the point is detected, the vehicle detection holding length L _C is set, and when the following relationship is established, the vehicle is rejected as not being the front edge 21 of the vehicle 21, and the D portion is mistaken for the vehicle front edge. It will not be detected.

│BC│ <Lc If the position of the front edge B of the vehicle 21 can be detected in this way, not only can accurate position information of the vehicle 21 be detected, but also Similarly, the difference from the position of the front edge portion B of the vehicle 21 which can be detected,
The speed of the vehicle 21 can be calculated . Also, CPU
The unit 19 processes the frame memory 15 to measure the luminance of the road surface, performs selection processing of parameters suitable for the road surface luminance, and controls the lens of the television camera device 11. The signal for driving the lens is generated by the lens control voltage generator 13.

FIG. 3 shows the relationship between road surface luminance and lens aperture control. When the road surface brightness becomes high and becomes equal to or higher than the close threshold value 31, that is, within the closed area 32, C
The PU unit 19 sends a lens aperture close command to the lens control voltage generation unit 13 to generate a lens aperture close signal. On the other hand, when the road surface luminance is reduced and becomes equal to or less than the open threshold value 33, that is, within the open area 34,
The CPU 19 sends a lens aperture open command to the lens control voltage generator 13 to generate a lens aperture open signal.
Send to In this way, the CPU unit 19 controls the television camera device 11 to maintain the road surface brightness in the road surface level holding area 35.
To control the lens.

As described above, according to the embodiment, high-speed imaging can be realized by reducing the number of effective scanning lines of the television camera device 11. In addition, vehicle position errors in screen imaging can be reduced, and most of the image processing such as frame difference and profile processing can be easily implemented in hardware. Can be
Highly accurate vehicle physical information can be detected.

[0019]

According to the present invention as described above, according to the present invention, the
Control the number of effective scanning lines of the TV camera device by the
Control to reduce errors in vehicle position detection.
Not only the current image data but also the previous image data
Data and subtract the pixels corresponding to the difference from the screen.
Profiles obtained by adding each in the horizontal and vertical directions
Detects vehicle width component and vehicle front edge from file data
Then, the position and speed of the vehicle can be calculated. Then, to achieve high-speed processing by dedicated hardware the major part of the image processing, high vehicle position detection accuracy, the speed measurement accuracy MotoSo
It can be realized only by the device. Therefore, the efficiency of vehicle detection and the improvement of economy can be improved. Ma
Further, the road surface luminance is calculated from the image data stored in the storage means.
The lens of the TV camera device according to the road surface brightness
By controlling the aperture, a good image can be captured.
Wear.

[Brief description of the drawings]

FIG. 1 is a schematic block diagram showing a vehicle detection device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram schematically showing data in a profile memory in the vehicle detection device.

FIG. 3 is an explanatory diagram of a lens control operation in the vehicle detection device.

[Explanation of symbols]

 Reference Signs List 11 TV camera device 12 Synchronization signal generation unit 13 Lens control voltage generation unit 14 AD conversion unit 15 Frame memory 16 Frame difference circuit 17 Profile circuit 18 Profile memory 19 Microprocessor unit

Claims (2)

(57) [Claims] 1. Synchronous signal generating means for supplying a synchronous signal
And the number of effective scanning lines can be arbitrarily changed by the above synchronization signal.
And the television camera device, the car picked up by the television camera
A storage unit for storing both image data, a subtraction process for subtracting the currently captured image data from the previous image data stored in the storage unit, and outputting a pixel corresponding to the difference. Processing means, and the difference pixel between the horizontal and vertical directions of the screen.
And flop <br/> profile processing means for obtaining profile data and addition processing respectively in countercurrent, drive from the profile data
By detecting both vehicle width components and the vehicle front edge , the vehicle
A vehicle detection device comprising: calculation means for measuring both positions and speeds . 2. The arithmetic means is stored in a storage means.
The road surface luminance is measured from the image data, and the text
The lens aperture of a Levi camera device is controlled.
The vehicle detection device according to claim 1.