JPS63220399A - Binary-coding system of image data for measuring traffic flow - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method for binarizing image data for measuring traffic flow, which measures traffic flow from image data obtained by an ITV camera, etc., particularly at night or inside a tunnel. It is used for image data obtained in dark places such as.

[Conventional technology]

Various image processing-based traffic flow measurement systems have been known for a long time, and a binarization method was also developed in Japanese Patent Application Laid-Open No. 1986-
Specific examples are shown in Publication No. 214100 and "Toshiba Review" Vol. 40 No. 8, P, P, 677-679.

By the way, in dark places such as roads at night or roads in tunnels, it is not easy to recognize the road surface or the vehicle body using image data. Conventionally, therefore, the headlights and taillights of a vehicle are detected, and based on this, the presence and running of the vehicle are recognized.

Hereinafter, the prior art will be explained based on FIG. 4 of the accompanying drawings.

FIG. 4 is a diagram for explaining the conventional binarization method,
Two images of the inside of the tunnel taken by the ITV camera are shown.

As shown in the figure, lanes 12A and 12B opposing each other in the tunnel are displayed on the screen 11.

Since the inside of the tunnel is normally dark, only the tunnel illumination lights 13, the vehicle headlights 14, and the taillights 15 have high brightness.

In such a screen 11, a measurement zone for measuring traffic flow is set as shown by reference numeral 16 (indicated by a broken line). Then, by binarizing the brightness of this measurement zone 16 at a predetermined threshold level, headlights 14 and tail lamps 15 are extracted.

[Problem that the invention seeks to solve]

However, the above conventional technology has the following problems. That is, since the brightness of the headlights 14 and the brightness of the tail lamps 15 are significantly different, errors occur in extraction when binarized at a fixed threshold level. especially,
There is no significant difference between the brightness of the road surface in front of the vehicle illuminated by the headlights 14 and the brightness of the tail lamps 15. As a result, the road surface in front of the headlights 14 may be mistakenly recognized as the taillights 15 of the vehicle.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a 2 flI conversion method for image data for traffic flow measurement, which can accurately extract feature points with different luminances within a screen.

[Means for solving problems]

The binarization method of image data for traffic flow measurement according to the present invention is as follows:
The image data obtained by photographing multiple lanes in a tunnel or by photographing multiple lanes at night is divided into at least two measurement zones corresponding to the multiple lanes, and each divided measurement zone is divided into a predetermined measurement zone. It is characterized in that it binarizes image data by setting a threshold level.

[Effect]

The binarization method of image data for traffic flow measurement according to the present invention is as follows:
Since the screen is configured as described above, the screen is divided into measurement zones with different brightness to be extracted, and each is binarized at a different threshold level. Therefore, in the measurement zone where the brightness of the extraction target is low, it is divided into measurement zones based on the low threshold level. In measurement zones with high brightness, extraction is performed based on a high threshold level.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. 1 to 3 of the accompanying drawings. In addition, in the description of the drawings, the same elements are given the same reference numerals, and redundant description will be omitted.

FIG. 1 is a diagram illustrating the setting of measurement zones according to the embodiment. The difference from the conventional example shown in FIG. 4 is that the measurement zone 16A is
, 16Blfi is set. Therefore, the tail lamp 15 of the vehicle will be photographed in the measurement zone 16A, and the headlight 14 of the vehicle will be photographed in the measurement zone 16B.

FIG. 2 is a basic configuration diagram of an image processing system to which the method of the present invention is applied.

As shown in FIG. 2, the image processing system includes a VTR 2 connected to a fixed camera 1, an image processing device 3 for processing image signals (image data) from the VTR 2, and a VTR 2 connected to a fixed camera 1.
It has a monitor television 4 for monitoring images from the computer, and a console 5 for operating the image processing device 3.

The image processing device 3 also includes an image memory 7 for digitizing the analog image signal from the VTR 2, a slave CPU 8 for digitizing the digital image signal, and a binary signal from the slave CPU 8 to control the vehicle. It has a master CPU 9 for recognizing the headlights 14 and taillights 15, etc. Furthermore, the VTR 2 not only receives image signals from the fixed camera 1 via the signal line, but also receives image signals from the portable camera 6 in the form of a videotape or the like.

Next, the operation of the above embodiment will be explained.

The analog image signal from the V-block R2 is converted into digital luminance data of, for example, 256 gradations by an A/D converter (not shown) built into the image memory 7. Then, binarization is performed using a predetermined threshold level.

FIG. 3 is a flowchart showing this binarization method. As shown in the figure, first, image data for each pixel is read (step 101), and it is determined whether all image data has been processed (step 102). If there is image data, it is the measurement zone 16.
It is determined whether it belongs to A (step 103).

This determination is made by comparing the read coordinate values of the pixels and the preset coordinate values of the measurement zone 16A.

When the read pixel is in the measurement zone 16A, binarization is performed using the threshold level α (step 104). That is, the luminance level F (D) of the pixel is α
When the level is above, the binarized data is set to "1", and when the brightness level F (p) of the pixel is at a level that does not reach α, the 21 normalized data is set to "Oop".
The value of this threshold level α is determined with reference to the brightness of the tail lamp 15 of the vehicle, so that when the tail lamp 15 is present, binary data of "1" is obtained.

After the above binarization is completed, the process returns to step 101 again to read the image data.

When it is determined in step 103 that the pixel is not in the measurement zone 16A, the process proceeds to step 105 and it is determined whether the pixel is in the measurement zone 16B.

Then, when it is determined that the pixel in the measurement zone 16B is at the threshold level β (α
Binarize the image data using β) (Step 1
06). That is, when the brightness level F (p) of the pixel is at a level higher than β, the binary data is set to “1”, and when the brightness level F (p) of the pixel is at a level that does not reach β, the binary data is set to “1”. Let be “Otp.In addition,
The value of this threshold level β is determined with reference to the brightness of the headlights 14 of the vehicle, so that when the headlights 14 are present, binary data of "1" is obtained.

After the above binarization is completed, the process returns to step 101 again to read the image data.

If it is determined in both steps 103 and 105 that the measured zone is not within the measurement zones 16A and 16B, no binarization is performed. Then, the above operations are repeated, and when there is no more image data to read, the binarization operation is ended (step 102).

In this way, in this embodiment, two measurement zones 16A, 1
Different threshold levels α and β are set for each 6B. Therefore, for the measurement zone 16A (low i1 degree zone) in which the tail lamp 15 of the vehicle is captured but the headlight 14 is not captured, the threshold level is set to a low level α, and the measurement is performed in which the tail light 15 of the vehicle is not captured but the headlight 14 is captured. Zone (high 8 degree zone) 1
For 6B, set the threshold level to a high level β
Therefore, even if features with different brightness appear on the same screen, no error will occur in the extraction.

The present invention is not limited to the above embodiments, and various modifications are possible.

For example, the number of measurement zones is not limited to two, but may be three or more. Also, the threshold level α.

The value of β may be appropriately corrected depending on environmental conditions (conditions such as brightness of the sky, strong lighting in the tunnel, etc.).

Furthermore, the binarization method is not limited to the one shown in FIG. 3, and any other method may be used. For example, if the entire screen is first binarized using a threshold level α, then the entire screen is binarized using a threshold level β.
The obtained binarized data may be divided into measurement zones 16A and 16B.

〔Effect of the invention〕

As explained above in detail, according to the binarization method according to the present invention, the screen is divided into zones with different brightness to be extracted, and each zone is binarized at a different threshold level, so that the brightness to be extracted is In low zones, extraction is performed based on a low threshold level, and in high brightness levels, extraction is performed based on a high threshold level, so even when there are feature points with different brightness in the screen, these features can be accurately extracted. There are effects that can be extracted.

[Brief explanation of the drawing]

FIG. 1 is a diagram for explaining measurement zone setting according to the embodiment method of the present invention, FIG. 2 is a basic block diagram of an image processing system to which the method of the present invention is applied, and FIG. 3 is an example of the embodiment of FIG. 1. FIG. 4 is a flowchart for explaining the binarization of image data according to the method, and FIG. 4 is a diagram for explaining the setting of measurement zones according to the conventional method. DESCRIPTION OF SYMBOLS 1...Fixed camera, 2...VTR, 3...Image processing device, 4...Monitor TV, 5...Operation console, 6...
...Mobile camera, 7...Image memory, 8...Slave CPU, 9...Master CPU, 11...Screen, 1
2A, 12B... Lane, 13... Tunnel lighting, 14... Headlight, 15... Tail lamp,
16.16A, 16B...Measurement zone. Patent applicant: Sumitomo Electric Industries, Ltd. Representative Patent Attorney Yoshiki Hase 16A
16B Setting of the measurement zone in the embodiment Fig. 1 Basic block of the device to which the method of the present invention is applied Fig. 2
Figure 4: Flowchart of binarization Measurement zone setting in conventional example

Claims (1)

[Claims] 1. Image data for traffic flow measurement in which image data obtained by photographing a plurality of lanes in a tunnel or by photographing a plurality of lanes at night is binarized at a predetermined threshold level.
In the value conversion method, the image data is divided into at least two measurement zones corresponding to the plurality of lanes, the predetermined threshold level is set to a different value for each divided measurement zone, and the image data is divided into at least two measurement zones corresponding to the plurality of lanes. A binarization method for image data for traffic flow measurement, characterized by binarization. 2. The binarization method of image data for traffic flow measurement according to claim 1, wherein the predetermined threshold level is changed in accordance with changes in environmental conditions.