JPH04263400A - Vehicle movement measuring unit - Google Patents

Abstract

PURPOSE:To accurately measure the position or speed of a vehicle by deciding the condition of a road surface, changing the updating rate for the road surface from the condition and always maintaining the correct background data. CONSTITUTION:A video camera means performs the image pickup of the movement of a vehicle on the road and stores the picture information once in a memory means. A picture processing means performs the processing for the stored picture information. The picture processing means changes the rate of the update of the background data by the decision method for the condition of the road surface and discriminates daytime, twilight and nighttime by the decision method for day and night. And this changes the threshold at the time of the picture processing and the updating rate of the background data by the background updating method. The update of the background data is performed accurately by the background difference system and the frame difference system and the extraction of the running vehicle is facilitated. The output can be utilized for the congestion information source or travelling time plan by transmitting the output to the central device.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] The present invention is installed on a road,
Vehicle speed, number of vehicles passing, vehicle type (normal vehicle, large vehicle, etc.)
This invention relates to a vehicle dynamic measurement device that collects necessary traffic information such as traffic information.

0002

[Prior Art] Conventionally, this type of vehicle dynamic measuring device performs image processing on current image data and background data photographing the movement of vehicles on the road, outputs the speed of the vehicle, the number of passing vehicles, and the vehicle type (regular car, It is configured so that it can output objects such as large vehicles.

0003

[Problems to be Solved by the Invention] However, in the conventional vehicle dynamic measurement device described above, the background data is always updated at a certain rate, so when the vehicle is stopped for a long time, such as during traffic jams, There was a problem that the background data could not be updated.

The present invention solves these conventional problems by determining the state of the road surface (such as the absence of a vehicle, the presence of a running vehicle, or the presence of a stopped vehicle), and It is an object of the present invention to provide a vehicle dynamics measuring device that can accurately measure the position and speed of a vehicle by changing the update rate of road surface data from this state and always maintaining correct background data.

[0005]

[Means for Solving the Problems] In order to achieve the above object, the present invention provides a video camera means for capturing an image of the movement of a vehicle on a road and outputting image data as an electric signal, and a video camera connected to the video camera means. and converting said image data from said video camera means into digital data.
D conversion means, input image memory means connected to the A/D conversion means and temporarily storing the digital image data, and background data connected to the A/D conversion means and used as a background when photographing a vehicle on the road. That is, background data memory means for storing road data when no vehicle is present on the road, the input image memory means, and the video camera means connected to the background data memory means and stored in these memory means. The background data obtained by
road surface conditions regarding the presence or absence of running vehicles and the presence or absence of stopped vehicles;
The apparatus includes an image processing means for determining the size of a traveling vehicle, and an output means connected to the image processing means for outputting the determination result.

[0006]

Therefore, according to the present invention, the road surface condition can be classified into the following states: no vehicle, running vehicle, and stopped vehicle based on the current image data and background data. In addition, when updating the background data, it is possible to update at a large update rate when there are no vehicles from the road surface information, and at a small update rate when there is traffic congestion. Therefore, accurate background data can be maintained and accurate It becomes possible to measure the vehicle's position and speed.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of the present invention will be described below with reference to the drawings.

(Embodiment 1) FIG. 1 is a block diagram showing the configuration of an embodiment of a vehicle dynamic measuring device according to the present invention.

In the figure, reference numeral 1 is placed so as to overlook the movement of vehicles on the road, and photographs the movement of vehicles on the road.
A video camera 2 serves as a video camera means for outputting image data as an electric signal, and is connected to the video camera 1, and is a vehicle dynamic measuring device body that determines road surface conditions based on information from the video camera 1. are its constituent elements. 3 is connected to the video camera 1 and serves as an A/D conversion means for converting image data output from the video camera 1 into digital data; 4 and 5 are connected to this A/D converter 3; An input image memory 6 serving as an input image memory means for temporarily storing an input image as digital data is also connected to the A/D converter 3, and is connected to a road with no vehicles photographed by the video camera 1. An image memory 7 serving as a background image memory means for temporarily storing background data, which is information, is connected to the A/D converter 3 and also to the image processing unit shown below, and is connected to the image processing unit 7 to store processed images processed by the image processing unit. A processed image memory 8 serving as a processed image memory means for temporarily storing images is connected to the memories 4 to 7, and performs image processing based on the input image and background image stored in the memories 4 to 6 to extract vehicles, The image processing unit is an image processing unit that evaluates tracking, driving speed, etc., determines the current road condition, and updates background data.

Next, the operation of the above embodiment will be explained. First, image information captured using the video camera 1 is sent to the vehicle dynamic measurement device main body 2. Vehicle dynamic measurement device 2
The A/D converter 3 converts the image information into digital data. Then, two screens worth of digital data captured at a certain interval are temporarily stored in the input image memories 4 and 5. Furthermore, information on road conditions where no vehicle is present, ie, background data, is temporarily stored in the background image memory 6. This background data changes depending on the time of day, such as morning, noon, and night, and the weather, such as sunny, cloudy, and rainy weather. Therefore, as shown in the conventional example, this background data is necessary to accurately determine the road surface condition. It is necessary to update the data according to the above conditions. Therefore, these input image memories 4
, 5 and the data in the background image memory 6, the processing result is written in the processed image memory 7, the vehicle is extracted from the image data, and this process is performed continuously. In addition to outputting vehicle tracking results and driving speed, it also determines the current road condition.
The background data is updated based on this information.

Next, the basic algorithm of the road surface condition determination method in the above embodiment will be explained with reference to the drawings.

[0012] Road surface condition divided into four stages from 0 to 3 (Table 1)
Shown below.

[0013]

[Table 1]

First, in step 2-1, a frame difference is performed between the input image memories 4 and 5 that store digital data for two screens taken at a certain interval. By performing this frame difference, it becomes possible to extract a moving object (in this case, a running vehicle). Therefore, as in step 2-2, by measuring the number of pixels that have changed in the frame difference, it is possible to distinguish whether there is a running vehicle, no vehicle, or a stopped vehicle. Can be done. Also, steps 2-3, 2-4, 2
-5, the size of the running vehicle is distinguished based on the size of the number of pixels that has changed in the frame difference. Next, in steps 2-6 and 2-7, the average brightness value and the most frequent brightness value of the input image memory 4 and the background image memory 6 are determined in order to distinguish whether there is a stopped vehicle. Further, the average luminance values are compared in steps 2-8 and 2-9, and if the difference is large, it is determined in step 2-11 that a stopped vehicle exists. Further, when the difference between the average luminance values is small and the difference between the most frequent luminance values is small, it is determined in step 2-10 that no vehicle is present. In this way, by comparing the most frequent luminance values, it is possible to determine the presence of a stopped vehicle even in cases where the average luminance value is small but a vehicle is present.

[0015] Thus, according to the above algorithm,
The background data update rate can be changed based on the road surface condition (no vehicle present - running vehicle present - stopped vehicle present). That is, the update rate is increased when no vehicle is present, and is decreased when a stopped vehicle is present, thereby making it possible to maintain more accurate background data.

Next, the processing procedure of the day/night determination method of the present invention will be explained with reference to FIG. In this case, as shown in Table 2, day and night are determined in three stages from 0 to 2.

[0017]

[Table 2]

First, in step 3-1, the average brightness of the background data in the background image memory 6 is determined. Next, in step 3-2, the day/night determination FLG is determined.

As shown in (Table 2), the day/night determination FLG
If it is 0, it is daytime, if it is 1, it is dusk, and if it is 2, it is night. Here, when the day/night determination FLG is 0, that is, daytime, in steps 3-3 and 3-4, if the average luminance of the background data is less than or equal to the daytime → twilight determination threshold α1 shown in the figure, the determination FLG is changed to twilight. Change to Further, when the determination FLG is 1, that is, twilight, in steps 3-5, 3-6, and 3-7, a threshold value based on the average brightness of the background data is used from the data in the input image memory 4 to The number of pixels is measured, and when the twilight→nighttime determination threshold α3 or more is reached, the determination FLG is changed to nighttime. In this way, in this embodiment, the headlight section is tracked at night, and it is checked whether or not to shift to night tracking processing based on whether there are many vehicles with their headlights on. Furthermore, when the day/night determination FLG is 2, that is, it is night, step 3-8
, in step 3-9, if the average luminance value of the background data is greater than or equal to the nighttime-to-daytime determination threshold α4, the determination FLG is changed to daytime.

As described above, according to the processing procedure of the day/night determination method of the present invention, it is possible to change the threshold value during image processing by determining daytime, twilight, and nighttime based on background data. In particular, when it is determined that it is twilight, and the difference in brightness between the vehicle body and the road surface in the current image data is small, the threshold value in image processing can be changed to a small value based on this information. Also, at night, the background data is basically stable, but when a vehicle passes by, the road surface may become bright due to the reflection of headlights, which may affect the update of the background data. . Therefore, if it is determined that it is nighttime based on the day/night determination information, the background data can be updated accurately by lowering the background data update rate.

Next, the processing procedure of the background updating method of the present invention will be explained with reference to FIG. First, in steps 4-1 and 4-2, the road surface condition and day/night are determined by the road surface condition determination subroutine and the day/night determination subroutine. In steps 4-3, 4-4, 4-5, 4-6, and 4-7, if the current state is night, the background is updated once every five times. Here, CNT in the figure is the update count number. Furthermore, in the daytime, the background update rate is changed depending on the road surface condition. That is, steps 4-8, 4
-9, if the road surface condition FLG shown in Table 1 is 0 (no vehicle exists), the background is always updated. Further, in steps 4-10 to 4-13, if the road surface condition FLG is 1 or 2 (that is, a running vehicle is present), the background is updated once every three times. Furthermore, steps 4-14 to 4-17
In the case where the road surface condition FLG is 3 (that is, there is a stopped vehicle), the background is updated once every 10 times.

As described above, according to the background updating method of the present invention, the update rate of background data can be lowered when the vehicle is stopped, such as during traffic jams, compared to the road surface information using the road surface condition determination method described above. This prevents the background data from collapsing, and increases the update rate when no vehicle is present, making it possible for the background data to sufficiently follow sudden changes in the surroundings. Additionally, by changing the background data update rate between day and night, the effects of road reflections from headlights can be avoided as much as possible at night.

Next, a processing procedure for determining the vehicle dynamics using the background difference method and frame difference method of the present invention will be explained with reference to FIG.

First, in step 5-1, initial background data is created and stored in the background image memory 6. Also,
In step 5-2, initial day/night determination processing is performed to determine whether it is day or night. Then, in step 5-3, frame difference processing,
In step 5-4, background difference processing is performed and processed data is stored in the processed image memory 7. The threshold values for these processes are changed depending on daytime, twilight, and nighttime. In particular, when it is determined that it is twilight, the vehicle portion can be easily extracted by reducing the brightness difference between the vehicle body and the road surface and reducing the threshold value. and,
In step 5-5, a vehicle candidate portion is extracted by determining the horizontal brightness distribution from this processed data. In step 5-6, this result is compared with the previous position of the vehicle candidate part, the distance traveled by the vehicle is measured, and the speed of the vehicle is measured from this result. Then, background data is updated in step 5-7, and day/night determination is performed in step 5-8.

As described above, according to the above processing procedure, the background data can be updated accurately based on the road surface information, and the background difference information becomes more accurate, making it easier to extract existing vehicles. . Further, by performing frame difference, it becomes easier to extract a running vehicle.

Next, referring to FIG. 6, a case will be described in which the vehicle dynamic state measuring device shown in FIG. 1 is provided with an output section. In the figure,
1 to 8 are exactly the same as in FIG. However, in the current case, an output section 9 for outputting the processing results is added. Then, the current degree of traffic congestion on the road is determined using the road surface condition determination method shown in FIG. The information is transmitted to the central device, and the central device can use the information as a source of information such as traffic jam information or to measure travel time.

[0027]

Effects of the Invention As is clear from the above embodiments, the present invention has the following effects.

1) The background data update rate can be changed depending on the road surface condition (no vehicle - running vehicle - stopped vehicle). That is, when there is no vehicle, the update rate is increased, and when a stopped vehicle is present, the update rate is decreased, making it possible to maintain more accurate background data. Furthermore, the degree of road congestion can be determined.

2) By determining daytime, dusk, and nighttime from the current image data, the threshold value for image processing can be changed. In particular, when it is determined that it is twilight, if the brightness difference between the vehicle body and the road surface in the current image data is small, the threshold value in image processing can be changed to a small value based on that information.

[0030] At night, the background data is basically stable, but when a vehicle passes by, the road surface may become bright due to the reflection of headlights, which affects the updating of the background data. Therefore, if it is determined that it is nighttime based on the day/night determination information, the background data can be updated accurately by lowering the background data update rate.

3) Based on the road surface information using the road surface condition determination method in 1) above, when the vehicle is stopped such as during traffic jams, the update rate of the background data is lowered to prevent the background data from collapsing. If the background data does not exist, increasing the update rate allows the background data to sufficiently follow sudden changes in the surroundings.

Furthermore, by changing the background data update rate between day and night, it is possible to avoid as much as possible the influence of headlights reflected on the road surface at night.

4) Background data can be updated accurately from road surface information, so background difference information becomes more accurate and existing vehicles can be easily extracted. Further, by performing frame difference, it becomes easier to extract a running vehicle.

5) By transmitting the information obtained in 1) to the central device, the central device can use the information as a source of information such as traffic congestion information and for travel time planning.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the vehicle dynamic measurement device of the present invention.

FIG. 2 is a flowchart showing the processing procedure of the road surface condition determination method in the embodiment of the present invention.

FIG. 3 is a flowchart showing the processing procedure of the day/night determination method in the embodiment of the present invention.

FIG. 4 is a flowchart showing the processing procedure of the background update method in the embodiment of the present invention.

FIG. 5 is a flowchart showing a processing procedure for determining vehicle dynamics using a background difference method and a frame difference method in an embodiment of the present invention.

FIG. 6 is a block diagram showing the configuration of a modification of the embodiment of the present invention.

[Explanation of symbols]

1 Video camera 2　Vehicle dynamic measurement device main body 3 A/D conversion section 4, 5 Input image memory 6 Background image memory 7 Processing image memory 8 Image processing section 9 Output section

Claims (5)

[Claims] 1. Video camera means for capturing an image of the movement of a vehicle on a road and outputting image data as an electrical signal; and a video camera connected to the video camera means for converting the image data from the video camera means into digital data. A to do
/D conversion means; input image memory means connected to the A/D conversion means and temporarily storing the digital image data; and a background connected to the A/D conversion means and used as a background when photographing a vehicle on the road. background data memory means for storing data, i.e. road data when no vehicle is present on the road; and said video camera means connected to said input image memory means and said background data memory means and stored in said memories. The obtained background data, the average brightness value of the current image data, the most frequent brightness value, and road surface information such as the presence ratio of moving objects are compared to determine the presence or absence of vehicles on the road, the presence or absence of running vehicles, and the presence or absence of stopped vehicles. 1. A vehicle dynamics measuring device comprising: an image processing means for determining road surface conditions, the size of a traveling vehicle, etc.; and an output means connected to the image processing means for outputting the determination results. 2. The vehicle dynamic measuring device according to claim 1, wherein the image processing means determines daytime, dusk, and nighttime based on the average luminance value of the background data, and changes a threshold value for image processing. 3. The vehicle dynamic measuring device according to claim 1, wherein said image processing means changes an update rate of background data based on said determined road surface condition and said day/night determination result. 4. The vehicle dynamics measuring device according to claim 1, wherein the image processing means collects the dynamics of the vehicle based on the difference result of subsequent background data and the difference result of the imaged screen frame. 5. The vehicle dynamic measurement according to claim 1, wherein the image processing means determines the degree of road congestion based on the difference result of the background data, the difference result between the captured two screen frames, and the determination result of the road surface condition. Device.