JPH05151495A - Car image processing system - Google Patents

Abstract

PURPOSE:To reduce cost by detecting a car with image processing in which brightness data sequentially obtained are compared with a reference level data at night and obtaining video signal, while lighted for a specified period, for car image processing. CONSTITUTION:When no car is present in a sensor area SA, defferential processing, for example, is performed to obtain brightness data DV based upon data (Dt) obtained from measuring points P1, P2... corresponding to the sensor area SA. If a car headlight is present in the sensor area SA, since the brightness data DV based upon the brightness data Dt obtained from measuring points P1, P2... corresponding to the sensor area SA contains the brightness data corresponding to car's headlight, it is of high level as a whole, having a peak corresponding to the headlight. Therefore, preset threshold level Ds is exceeded, car detection output thus obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle image processing method for picking up an image of a vehicle with a video camera and processing the obtained video signal to detect the vehicle. The present invention relates to a vehicle image processing method.

[0002]

2. Description of the Related Art Image processing for obtaining a video signal by an image pickup means and processing the signal to identify an object to be imaged and to determine a change in the imaged object is often used for traffic control. For traffic control, for example, as shown in FIG. 7, a road is photographed using a video camera 10 installed approximately 6 m above the road side, and the image is processed by a host device (image processing device) not shown. An image processing device for detecting the traveling vehicle CA and recognizing the registration number of the license plate has been used, and has been successful in grasping traffic volume and identifying speeding vehicles. Such a device has been proposed by the applicant as Japanese Patent Application No. 2-318992 (image type vehicle detector), for example, and a vehicle image processing method is also mentioned.

Needless to say, this kind of device is operating day and night, and in order to obtain a clear image signal suitable for image processing even at night when the illuminance is insufficient, for example, infrared rays which are invisible light are used. The illuminating device used for image capturing is used. As shown in FIG. 7, the lighting device 50 is attached integrally with the video camera 10 to illuminate the image pickup area. The above-mentioned lighting device does not always operate at night when there are few moving vehicles because the service life is extended or repair intervals are extended. ing. Conventionally, in order to detect a traveling vehicle to start this lighting operation, an appropriate vehicle detection device such as a loop coil type vehicle detector and an ultrasonic type vehicle detector is separately provided at an upstream position on the road away from the image pickup point and imaged. The approach of a traveling vehicle to the point was detected.

[0004]

It is necessary to separately provide the vehicle detection device at night and drive the illumination device for the image processing device based on the output thereof as in the conventional case, in addition to the cost of the device and the execution cost thereof. The cost burden was heavy. The present invention detects a vehicle at night and drives the lighting device for a predetermined time without separately using a vehicle detection device as in the prior art, and thus does not propose a low cost vehicle image processing method. It is what

[0005]

In order to solve the above problems, according to the present invention, a video signal corresponding to a region including a road to be sensed is obtained, and a plurality of measurement points on the screen are timed in the video signal. Luminance data group (D
In the vehicle image processing method for sequentially obtaining t) and performing predetermined image processing based on these to perform vehicle detection and the like, in the night, a luminance data group (Dv) based on the sequentially obtained luminance data group (Dt) The vehicle is detected by the image processing for comparing with the reference level data group (Dr), illumination is performed for a predetermined time, and a video signal is obtained and predetermined vehicle image processing is performed.

[0006]

The signal for starting the driving of the lighting device at night is obtained by image processing similar to the original vehicle image processing.
Therefore, no separate device for vehicle detection is required.

[0007]

The present invention will be described in detail below with reference to the accompanying drawings. For vehicle image processing, for example, an image type vehicle detection device as shown in FIG. 3 is used. The image-type vehicle detection device is roughly composed of an image pickup means 10, a brightness data conversion section 20, an arithmetic processing section 30, a control section 40, and a night illumination device 50. The image pickup means 10 is, for example, a CCD video camera, and picks up an image of the detection area of the road to correspond to the NTSC as shown in FIG. 4A or 4A ′.
The system video signal is sent to the luminance data converter 20.

The brightness data converter 20 is a part for converting a video signal into digital data. A sync signal extraction circuit 21 for obtaining a horizontal sync signal and a vertical sync signal from the video signal, and a current sync line extraction circuit based on these sync signals. A horizontal address counter 22 and a vertical address counter 23 for digitizing the investigation position to obtain corresponding horizontal coordinates and vertical coordinates. The moment when these outputs are compared with the coordinates sequentially designated by the microprocessor 41 (described later) and coincident with each other. And a video A / D conversion circuit 25 for digitizing the luminance of the video signal at a corresponding moment according to the captured signal to obtain the luminance data and outputting the digitized luminance data. The input section has a video amplifier 27 that amplifies the transmitted video signal to an appropriate level.
A clamp level fixing circuit 28 for normalizing the video signal to a constant amplitude prior to the / D conversion is also provided. The arithmetic processing unit 30 is a circuit for receiving the luminance data from the video A / D conversion circuit 25 and averaging a plurality of luminance data and calculating the correlation. The processor 31 is used and a storage unit necessary for storing data is also provided.

The control unit 40 controls each of the above-mentioned units, receives the data from the arithmetic processing unit 30 and further performs arithmetic processing on the data to detect the presence of a vehicle and calculate the speed of the vehicle. The part that outputs to the device, the microprocessor 41, the ROM that stores the control program
42, a RAM 43 for data storage, and an I / O circuit 44 for output. Reference numeral 45 denotes a CPU bus through which the microprocessor 41 exchanges data and addresses, control commands and response signals with other parts.

The illumination device 50 is driven by the I / O circuit 44 when the control section 40 detects a vehicle by image processing at night, and illuminates the image pickup target area for a predetermined time to obtain an image suitable for image processing. belongs to. This lighting device 50
For example, an infrared light emitter can be used. Further, the illumination device 50 may intermittently repeat the light emission illumination in synchronization with the image pickup operation (shutter timing) during a predetermined time, or may continuously emit light. In addition to the above, the example device also includes a DC power supply 51, a clock circuit 52 supplied to each unit, and a signal for increasing the sensitivity of the image pickup unit 10 when the level of a video signal from the image pickup unit 10 is low and inappropriate for processing. A D / A conversion circuit 53 for transmitting the signal is provided.

Vehicle image processing for processing a video signal using such a device to detect a traveling vehicle and identify a vehicle registration number will be briefly described. A plurality of measurement points (P ₁ , P ₂ , set on a straight line (detection line) assumed to cross the road in the image as shown in FIG.
...) and the luminance level (X ₁ , X ₂ , ...) Of the video signal temporally corresponding to each other, the luminance data group (Dt).
Are sequentially obtained at predetermined time intervals. Image processing for detecting the presence of a vehicle on the detection line at a certain time is performed based on the difference between the obtained brightness data groups. That is, the difference between the data group (threshold level data, Ds) determined in advance to cope with the case where no vehicle exists on the detection line and the brightness data group (vehicle detection level data) on the detection line is detected. The vehicle is detected accordingly. Alternatively, a group of brightness data when there is no vehicle on the detection line (reference road surface level data, Dr) and brightness data when there is a vehicle on the detection line and the brightness level changes significantly along the detection line. The vehicle is detected by detecting the difference from the group (vehicle detection level data). Further, if a plurality of detection lines are set, the speed of the vehicle can be calculated based on the arrival time of the vehicle to each detection line. Further, if the detection line is set on the entire screen and more brightness data is closely captured, the entire screen can be grasped, and the license plate can be identified. If necessary in consideration of the processing speed, an image at a certain point may be temporarily stored as a still image and the stored image may be processed.

As described above, the vehicle image processing method generally includes the following steps as shown in the flowchart of FIG. That is, 1) A video signal corresponding to an image of a region including a road to be sensed is obtained by an image pickup means such as a CCD camera. 2) A straight line (detection line) across the road in the video signal
A plurality of positions (temporal positions in the video signal) corresponding to the measurement points located above are set in advance, and luminance data groups (Dt) corresponding to these positions are sequentially obtained. 3) The luminance data group (Dt) along the detection line is set to the threshold level (Ds) or the reference road surface level (D).
Image processing such as detecting a vehicle by comparing with r) or calculating the level difference between the brightness data groups to identify the vehicle license plate and its vehicle registration number. Even in the method of the present invention, image processing is performed based on the above-described process during the daytime when the illumination device is not required, as in the conventional case.

By the way, in the method of the present invention, at night when illumination is required, for example, the vehicle is first detected by using the above-mentioned device as a part of image processing, and thereby the drive control of the illumination device is performed. After that, the original image processing is performed. In the vehicle image processing method of the present invention, vehicle detection at night is performed by monitoring the light from the headlights of the vehicle. That is, when no vehicle (headlight) is present in the sensor area SA as shown in the image of FIG.
Measurement points (P ₁ , P ₂ , ...) Corresponding to the sensor area SA
A luminance data group (Dv) is obtained by, for example, differentiating the data group (Dt) obtained from the above. As shown in FIG. 4B, the brightness data group (Dv) does not include brightness data corresponding to the headlights of the vehicle, and thus has a low level and is relatively uniform. Therefore, the threshold level (Ds) set in advance is not exceeded, and the vehicle detection output cannot be obtained.

Next, when the vehicle headlight exists in the sensor area SA as shown in the image in FIG. 4 (a '), the measurement points (P ₁ , P ₁₎ corresponding to the sensor area SA.
₂ , the luminance data group (Dv) based on the luminance data group (Dt) obtained from the luminance data group includes high-level luminance data corresponding to the headlights of the vehicle as shown in FIG. 5 (b '). Peaks corresponding to headlights (two in the figure)
have. Therefore, the preset threshold level (Ds) is exceeded, and the vehicle detection output is obtained. As a result, the driving of the lighting device is started and continues for a certain period of time. Alternatively, the drive may be stopped by obtaining a signal indicating no vehicle detection. Instead of the above threshold level (Ds), a brightness data group (Db, referred to as a reference background level or a reference road surface level) obtained when no vehicle is present can be used. The point is that the preset reference level data group (Dr) and the luminance data group (Dt) are compared to detect the presence or absence of headlights, that is, the presence of a vehicle. FIG. 1 is a system flow chart showing an embodiment of the process at night of the method of the present invention.
2 is divided and shown in FIG.

In the above process, the luminance data group (D
The luminance data group (Dv) based on t) is suitable for processing, for example, by computing adjacent values and applying a differential mask (differential processing), or averaging corresponding data of a plurality of measurement lines. It is intended to be converted to data.
However, (Dt) can be used as it is, and in this case, (Dv = Dt) is also included. Here, an example of the differential processing will be described. FIG. 5A shows the brightness levels (X ₁ , X ₂ , corresponding to each measurement point included in one measurement line on the screen.
Is a graph along the measurement line (in the road width direction). From each value of this brightness level, the differential value of the brightness data corresponding to a certain measurement point can be obtained by calculation. That is, if the differential value X'n corresponding to the measurement point Pn of FIG. 5A is to be obtained (an example of applying a 5 × 1 differential mask), the following equation, differential value X'n of the measurement point Pn = ( -1) X _n-1 + (-2) X _n-1 + 6X _n + (-2) X _{n + 1} + (-1) X _{n + 2} (where X _n-2 , X _n-1 , X _n, ... to obtain a differential value X'n performs arithmetic according luminance level) of each measurement point. Figure 5
A graph of a differential value obtained by applying a (5 × 1) differential mask to the luminance data group (Dt) in (a) (luminance data group, D
v) is shown in FIG. As can be seen from the figure, the differential value of the measurement point (for example, Pq) in which the brightness level is rapidly increasing from the adjacent measurement point is outstanding. As a result, a detection target such as a vehicle can be detected with a good S / N ratio. This differentiating process also gives good results for other image processing such as identifying the registration number of the license plate.

As described above, in the vehicle image processing method of the present invention, the image processing is performed at night based on the following processes including the detection of the vehicle. 1) An image pickup unit such as a CCD camera obtains a video signal corresponding to an image of a region including a road to be sensed. 2) In the video signal, a plurality of positions (temporal positions in the video signal) corresponding to measurement points located on the sensor region in the transverse direction of the road are set in advance, and the luminance data group (Dt) corresponding to these positions is set. Sequentially. 3) Luminance data group (D) based on the luminance data group (Dt)
The vehicle is detected by comparing v) with a reference level data group (Dr) such as a threshold level (Ds) or a reference road surface level (Db). 4) Drive the lighting device to start lighting. 5) A new luminance data group (Dt) is sequentially obtained. 6) The luminance data group (Dt) or the converted luminance data group (Dv) is used as a reference level data group (D) such as a threshold level (Ds) or a reference road surface level (Db).
R) is used to detect the vehicle, and the level difference between the luminance data groups is calculated to perform predetermined image processing such as detecting the vehicle license plate and identifying the vehicle number. 7) When there is no vehicle to be processed, the driving of the lighting device is stopped.

In short, in the method of the present invention, the illuminating device is not normally operated at night, and the vehicle is detected by image processing by focusing on the headlight of the vehicle, and the illuminating device is driven for a predetermined time by this detection output. At the same time, the original image processing such as identification of the target vehicle number is performed. For this reason, the conventional detection device used only for vehicle detection is unnecessary, and vehicle image processing can be performed at low cost. It is needless to say that the method of the present invention can be applied to both the case of using the pulse lighting type night lighting device and the continuous light emitting type night lighting device.

[0018]

As described above in detail, according to the method of the present invention, at night, the luminance data group (Dv) based on the sequentially obtained luminance data group (Dt) is changed to the reference level data group (Dt).
The night illumination device that contributes to the long life of the illumination device and the saving of power because the vehicle is detected by the image processing compared with r), the illumination is performed for a predetermined time, and the predetermined vehicle image processing is performed by obtaining the video signal. Control of the vehicle can be performed without the need for a separate device for vehicle detection, which makes the vehicle image processing device inexpensive, and the facility cost of the separate device can be eliminated and vehicle image processing can be realized at low cost. Very useful.

[Brief description of drawings]

FIG. 1 is a flowchart partially showing an embodiment of a vehicle image processing method of the present invention.

FIG. 2 is a flowchart partially showing an embodiment of the vehicle image processing method of the present invention.

FIG. 3 is a block diagram showing an example of an image type vehicle detection device according to the present invention.

FIG. 4 is a diagram illustrating an image and corresponding brightness data of the image type vehicle detection device.

FIG. 5 is an explanatory diagram of a differential value of brightness level data according to the present invention.

FIG. 6 is a flowchart showing an example of the daytime of the vehicle image processing method of the present invention.

FIG. 7 is a diagram for explaining a vehicle detection process and the like of the image type vehicle detection method.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Imaging means, 20 ... Luminance data conversion part, 30 ... Arithmetic processing part, 40 ... Control part, 50 ... Illumination device.

Claims (1)

[Claims] 1. A brightness data group (D) at a time position corresponding to a plurality of measurement points in the screen in the video signal is obtained by obtaining a video signal corresponding to a region including a road to be sensed.
In the vehicle image processing method for sequentially obtaining t), performing predetermined image processing based on these, and performing vehicle detection, the luminance data group (Dv) based on the luminance data group (Dt) sequentially obtained at night. The vehicle image processing method is characterized by detecting a vehicle by an image processing for comparing with a reference level data group (Dr), performing illumination for a predetermined time, and obtaining a video signal to perform a predetermined vehicle image processing.