JP2853361B2 - Vehicle dynamics measurement 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 dynamics measuring device which is installed on a road and measures and collects necessary traffic information such as vehicle speed, number of vehicles passing, and type of vehicle (normal vehicle / large vehicle).

[0002]

2. Description of the Related Art Conventionally, a vehicle dynamics measuring device of this type performs image processing of original image data and background data obtained by photographing a trend of a vehicle on a road to obtain the speed, the number of vehicles passing, and the vehicle type (normal vehicle / It is configured to measure and collect large vehicles).

[0003]

However, in the above-mentioned conventional vehicle dynamics measuring device, when performing image processing, the shadow of a vehicle in an adjacent lane is determined to be a vehicle and may be erroneously detected. In some cases, the shadow portion was determined to be the leading edge of the vehicle.

[0004] Further, there is another problem that when the luminance is low at dusk or the like, it cannot be detected when the luminance difference from the road surface of a black vehicle or the like is small.

The present invention has been made to solve the above-mentioned conventional problems, and has an accurate vehicle speed, the number of vehicles passing therethrough, and the type of vehicle (regular vehicle / vehicle) without being affected by the shadow of a vehicle in the front of the vehicle or in an adjacent lane. It is an object of the present invention to provide an excellent vehicle dynamics measurement device capable of measuring and collecting necessary traffic information such as a large vehicle.

[0006]

In order to achieve the above object, the present invention uses a video camera to process image data obtained by photographing the movement of a vehicle on a road and to measure and collect the dynamics of the vehicle. In a vehicle dynamics measurement device, when there is a shadow of a vehicle in front of the vehicle or in an adjacent lane in the daytime, the influence of the shadow is removed by using only the positive components of the background difference method and the frame difference method with expansion processing. Things.

In addition, when the luminance difference at dusk is small, the original image data and the background data are compared, and a threshold value is selected in two steps for each pixel, and only the plus components of the background difference method and the frame difference method with expansion processing are used. This removes the influence of shadows.

[0008]

The present invention has the following functions by the above-mentioned structure.

First, the vehicle can be tracked and measured without being shaded by a vehicle running in front of the vehicle or in an adjacent lane, and accurate measurement and collection of traffic information can be performed. Secondly, a black vehicle or the like having a small difference in luminance from the road surface at twilight can be extracted and the vehicle can be tracked and measured, and accurate traffic information can be collected.

[0010]

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

FIG. 1 shows the configuration of the embodiment. In FIG. 1, reference numeral 1 denotes a video camera, and 2 denotes a vehicle dynamics measurement device main body.

In the vehicle dynamics measuring device main body 2, 9 is A
/ D converter, 10 is an image memory (input image 1), 11 is an image memory (input image 2), 12 is an image memory (input image 3), 13 is an image memory (input image 4), 14 is image data processing Reference numeral 15 denotes a data output unit.

Next, the operation of the above configuration will be described. Video information captured using the video camera 1 is transmitted to the vehicle dynamics measurement device main body 2.

The vehicle dynamics measuring device main body 2 converts this information into digital data using an A / D converter 9. Then, digital data for two screens imaged at certain intervals are stored in the image memories 10 and 11. The image memory 12 stores information (background data) indicating that no vehicle exists.

The image memories 10, 11, 1
The processing result is written in the image memory 13 by using only the plus components of the background difference method and the frame difference method with expansion processing in the image data processing unit 14 using the data of No. 2, and the vehicle is extracted from the image data. By continuously performing this processing, the tracking and running speed of the vehicle are output by the data output unit 15, the current state on the road is determined, and the background data is updated based on this information.

FIG. 2 shows a case where only the plus components of the frame difference and the background difference with expansion processing are used.

First, t-.DELTA.t [se] shown in FIG.
c], image data of t [sec] shown in FIG. 2B, and background data shown in FIG. 2C.

First, t-Δt [sec], t [sec]
Of the image data. This is called a frame difference, and if only the plus component is made valid, the front edge of the traveling vehicle and the rear portion of the shadow beside the vehicle can be extracted as shown in FIG. Since the moving object pair can be reliably extracted from the frame difference, the expansion processing is performed rearward (the hatched portion in FIG. 2D), so that the difference becomes more reliable.

Further, the difference between the image data at t [sec] and the background data is calculated. This is called a background difference, and if only the plus component is made effective, the shadow portion is not extracted, and only the bright portion of the vehicle can be extracted as shown in FIG.

Then, by obtaining the logical sum of the frame difference and the background difference to create a processing screen, the vehicle can be reliably extracted as shown in FIG. For a vehicle of a bright white system, extraction is performed from both the frame difference and the background difference.

[0021] In the case of a vehicle of a dark black system, the brightness increases basically as the vehicle becomes closer to the image, so that the vehicle can be extracted by the frame difference.
A stopped vehicle cannot be extracted by the frame difference, but is extracted by the background difference.

Next, the process of removing shadows during the daytime will be described. FIG. 3 shows a flowchart of a processing screen creation process in the daytime with the shadow removed. First, image data t [s
ec] and t−Δt [sec] (frame difference), and only the plus component is made valid (step (S) 3).
1). Subsequently, a difference (background difference) between the image data t [sec] and the background data is performed, and only the plus component is made valid (step 32).

Then, first, the data of each pixel of the frame difference is compared with the threshold value α1 (step 35). And
In the case of Yes in the case of the threshold value α1 or more, “1” is written into the same position on the processing screen and the same column one or two rows before, and expansion processing is performed (step 36). If the data of each pixel of the frame difference is No equal to or smaller than the threshold value α1, the image data of the background difference is compared with the threshold value α2 (step 37). If the answer is Yes, which is equal to or larger than the threshold value α2, “1” is written at the same position on the processing screen (step 38). Otherwise, "0" is written at the same position on the processing screen (step 39). This is processed for all pixels to create a processed screen (steps 33, 34, 40, 41, 42, 43).

As described above, by using the processing scene, the shadow of the vehicle in the adjacent lane is not determined and tracked, and the leading edge position of the vehicle can be tracked accurately, so that accurate traffic information can be obtained. Can be measured and collected.

Hereinafter, processing for creating a processed screen at twilight will be described. FIG. 4 shows a flowchart of the processing screen creation process at twilight.

First, the image data t [sec] and t-Δt
A difference (frame difference) of [sec] is performed, and only the plus component is made valid (step (S) 51). Subsequently, the difference between the image data t [sec] and the background data (background difference)
And make only the plus component valid (step 5
2). Then, the image data t [sec] is compared with the background data for each pixel (step 55). If the difference is smaller than the threshold, Yes, the threshold of the frame difference and the background difference is １ ／ of the normal threshold (step 56). Otherwise, a normal threshold is used (step 57). By this processing, a vehicle with a small difference in luminance can be extracted.

Then, the data of each pixel of the frame difference is compared with the threshold value α1 (step 58). If the answer is Yes, which is equal to or greater than the threshold value α1, “1” is written into the same position on the processed screen and the same column one or two rows before, and expansion processing is performed (step 59). If the data of each pixel of the frame difference is equal to or smaller than the threshold α1, the data of each pixel of the background difference is compared with the threshold α2 (step 60). If it is equal to or larger than the threshold value α2, “1” is written at the same position on the processing screen (step 61). Otherwise, "0" is written at the same position on the processing screen (step 62). This is processed for all pixels to create a processed screen (steps 53, 54, 63,
64, 65, 66).

In this manner, a black vehicle or the like having a small difference in luminance from the road surface at twilight can be extracted from the processing screen, and accurate vehicle tracking and measurement can be performed, and accurate traffic information can be measured and collected. Can be.

[0029]

As apparent from the above embodiment, the present invention has the following effects.

It is possible to track the leading edge position of the vehicle without determining the shadow of the vehicle in the adjacent lane as the vehicle, and to measure and collect accurate traffic information.

In addition, a black vehicle or the like having a small difference in luminance from the road surface at twilight can be extracted, and the vehicle can be tracked and measured, so that accurate traffic information can be measured and collected.

[Brief description of the drawings]

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

2A is an image data diagram of t−Δt [sec] in the embodiment; FIG. 2B is an image data diagram of t [sec] in the embodiment; FIG. 2C is a background image data diagram of the embodiment; Image data showing the frame difference with expansion processing using only the plus component. (E) Image data showing the background difference using only the plus component in the embodiment. (F) Image data showing the background difference + frame difference with the expansion processing in the embodiment.

FIG. 3 is a flowchart of a processing screen creation process in the daytime in which a shadow has been removed in the embodiment.

FIG. 4 is a flowchart of a processing screen creation process at twilight in the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Video camera 2 Vehicle dynamics measuring device main body 9 A / D conversion part 10, 11, 12, 14 Image memory 14 Image data processing part 15 Data output part

Claims (2)

(57) [Claims] 1. A car traveling on a road using a video camera
By capturing both trends and processing the captured data
Measurement and measurement of vehicle dynamics such as speed and acceleration and vehicle type.
In the vehicle dynamics measurement device to collect and collect information on the vehicle information measurement area,
And provide the original data from the image data in the vehicle information measurement area.
Means for extracting image data and background data, and means for extracting a frame difference between the extracted original image data
Stage and means for expanding and extracting the data of the frame difference
And a difference between the original image data and the background data as a background difference
Means for extracting and extracting the positive component of the expanded frame difference and the background
The image data captured using the positive component of the difference is analyzed and processed.
And means for analyzing the image data to determine an adjacent lane and the same lane.
Removes the shadow component of the car on the line, eliminating measurement errors due to shadows
Vehicle dynamics measuring device characterized by being able to do . 2. A car traveling on a road using a video camera.
By capturing both trends and processing the captured data
Measurement and measurement of vehicle dynamics such as speed and acceleration and vehicle type.
In the vehicle dynamics measurement device to collect and collect information on the vehicle information measurement area,
And provide the original data from the image data in the vehicle information measurement area.
Image data and background data are extracted, and the background data
Calculate the average luminance value and recognize dusk based on the average luminance value.
Means, when the twilight recognition means is recognized as twilight,
The original image data is compared with the background data, and the threshold is determined in pixel units.
Means for sorting and processing the values in two stages.
Due to this, even at dusk, the shadow of the adjacent lane and the car in the same lane
The vehicle dynamics measurement device according to claim 1, wherein the influence can be removed .