JP2003296878A - Vehicle detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle detection method capable of improving the detection accuracy of a vehicle position. <P>SOLUTION: An image processing part 23 determines whether or not the surface reflection of a road 2 is present from the pixels of an image obtained by photographing a vehicle 3 traveling on the road 2. When it is determined that the surface reflection of the road 2 is present, a vehicle detection part 24 adds luminance in a horizontal direction for the respective pixels of a processed image in which the image is edge-processed and detects the position of the vehicle on the basis of the added result. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle detection method for photographing a vehicle traveling on a road and detecting the position of the vehicle.

[0002]

2. Description of the Related Art Conventionally, as a vehicle detection method, for example,
A background subtraction method is known in which an image obtained by photographing with a photographing means such as a CCD camera is compared with an image obtained when the vehicle is not passing to detect the position of the vehicle. With this method, the position of the vehicle can be detected from the image obtained by the photographing means.

[0003]

However, in the conventional vehicle detection method, there is a possibility that the reflected image of the vehicle, which is reflected on the road surface in rainy weather, is detected as the vehicle, and the position of the vehicle cannot be accurately detected. is there.

The present invention has been made to solve the above problems, and an object of the present invention is to provide a vehicle detection method capable of improving the detection accuracy of the vehicle position.

[0005]

In order to achieve such an object, a vehicle detection method according to the present invention has a road surface reflection of a road based on the brightness in pixels of an image obtained by photographing a vehicle traveling on the road. A road surface determination step of determining whether or not
When it is determined that there is road surface reflection, an addition step of adding the luminance in the horizontal direction to each pixel of the processed image obtained by edge processing the image, and a detection step of detecting the position of the vehicle based on the result of the addition. It is characterized by having.

Further, in the vehicle detecting method according to the present invention, in the detecting step, after detecting the first peak and the second peak for each added value of the horizontal luminance added in the adding step,
The position of the minimum value is detected between the position of the first peak and the position of the second peak, and the position of the vehicle is detected with the position of the minimum value as the head portion of the vehicle. .

According to the present invention, when it is determined that there is road surface reflection, the horizontal luminance is added to each pixel of the processed image obtained by edge processing the image. The value obtained by adding the luminance in the lateral direction tends to be large (has a peak) at the position corresponding to the vehicle and its reflection image. On the contrary, at the position corresponding to the boundary between the vehicle and its reflection image, this value tends to be relatively smaller than the value at the peak. By utilizing these tendencies, the boundary position between the vehicle and the reflected image can be accurately determined. Therefore, the position of the vehicle can be accurately detected in the photographed image even in rainy weather.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the accompanying drawings. In each figure, the same elements are designated by the same reference numerals, and overlapping description will be omitted. Further, the dimensional ratios in the drawings do not always match those described.

A vehicle detection method according to the embodiment will be described.

FIG. 1 is a block diagram of a vehicle detection device used for carrying out the vehicle detection method according to the present embodiment. As shown in FIG. 1, the vehicle detection device 1 detects the position of the vehicle based on an image obtained by capturing an image of the vehicle 3 traveling on the road 2. For example, the vehicle detection device 1 may include information indicating the detected vehicle position. It is used for collision prevention support and the like for informing the vehicle 4. This vehicle detection device 1 is a CCD (Charge Coupled Devic
e) A camera 10 and an image processing device 20 are provided.

The CCD camera 10 is for photographing the vehicle 3, and is fixed, for example, diagonally above the road 2,
The vehicle 3 traveling in the direction is photographed from diagonally above the front or the rear. The CCD camera 10 is an imaging camera capable of shooting a moving image, but may be a still image.

The vehicle detection device 1 further includes an image processing device 20. The image processing device 20 is connected to the CCD camera 10, inputs a captured image from the CCD camera 10, and detects the position of the vehicle using the captured image in the captured image.

The image processing apparatus 20 has an A / D (analogue-t).
An o-digital) conversion unit 21, an image memory 22, an image processing unit 23, and a vehicle detection unit 24. A
The / D conversion unit 21 performs A / D conversion of an input image signal or the like from an analog signal to a digital signal, and for example, a video A / D converter or the like is used. In addition,
When the camera 10 outputs a digital signal, the installation of the A / D converter 21 may be omitted.

The image memory 22 stores the digitally converted image signal as image data. For example, the image data is a set of pixels having a gray value for each coordinate point on the x axis and the y axis, and the gray level of the pixel is, for example, 256 gradations of 8 bits.

The image processing unit 23 performs image processing on the image data stored in the image memory 22. The vehicle detection unit 24 detects the position of the vehicle based on the information processed by the image processing unit 23.

Next, the operation of the vehicle detection device 1 and the vehicle detection method will be described in detail.

First, the vehicle 3 running on the road 2 is CC
The image is taken by the D camera 10. At this time, as shown in FIG. 2, when the road surface of the road 2 is wet due to rain or the like, the vehicle 3 is reflected on the road surface. In the captured image 30 captured by the CCD camera 10, both the vehicle 3 and the reflection image 3a reflected on the road surface are shown. Then, the captured image 30 including the reflected image 3a is input to the image processing device 20 as an image signal.

The image signal input to the image processing device 20 is A / D converted by the A / D converter 21 from an analog signal to a digital signal. The digitally converted image signal is stored in the image memory 22 as image data (luminance value of each pixel).

Then, according to the flow chart shown in FIG. 3, the following processing is performed by the image processing apparatus 20. That is, in S1 of FIG. 3, the image data stored in the image memory 22 is read by the image processing unit 23. The data read at this time is, for example, brightness data of the basic image data at the time of fine weather when the vehicle 3 is not photographed and the image data photographed this time.

Then, the flow shifts to S2, and the road surface determination processing is performed by the image processing unit 23. Road surface judgment processing is performed on road 2
It is a process of determining whether or not there is road surface reflection of the road 2 based on the brightness in the pixels of the image of the vehicle 3 traveling in the direction. Here, for example, the luminance values of the basic image data (data in fine weather) and the image data captured this time are compared, and when the difference between the luminance values is equal to or more than a predetermined value, it is determined that there is road surface reflection.

If it is determined in S2 that there is no road surface reflection, the process proceeds to S7 and vehicle detection is performed by background subtraction processing and edge processing. The background subtraction process is a process of subtracting a background image stored in advance from the captured image 30, and the edge process is a process of extracting an edge portion in which the brightness value of each pixel of the subtracted image greatly changes. The vehicle is detected by this extraction.

On the other hand, if it is determined in S2 that there is road reflection, the process proceeds to S3, and the image processing unit 23 performs the clipping process. The cutout process is a captured image 3 including the vehicle 3.
This is a process of cutting out a partial area 30a of 0 as a cut-out image 40. FIG. 4 shows a cutout image 40 cut out by the cutout processing. The cutout portion is determined by, for example, the background difference method. The cutout image 40 is
The coordinates are converted so that the width direction and the length direction of the road 2 are the x axis and the y axis of the image.

Then, the process shifts to S4 in FIG. 3, and the image processing unit 23 performs edge processing. The edge process is a process of detecting and extracting an edge portion in which the luminance value of each pixel of the cutout image 40 in FIG. 4 largely changes. By this edge processing, the contours of the vehicle 3 and its reflection image 3a are made clear on the cut-out image 40. In addition, this edge processing makes clear the edge portion where the brightness changes greatly in the vehicle 3 and its reflection image 3a. The processed image subjected to this edge processing is shown in FIG.

Then, the process shifts to S5 in FIG. 3, and the addition processing is performed by the vehicle detection unit 24. Here, the addition process is a process of adding the luminance in the horizontal direction, that is, the x-axis direction, to each pixel of the processed image obtained by edge-processing the image, as shown in FIG.

Then, the process shifts to S6 in FIG. 3 and the vehicle detection unit 24 performs a detection process. Here, the detection process is a process of detecting the position of the vehicle 3 based on the result of the addition. Here, the position of the vehicle 3 is detected by analyzing the feature of each added value to which the brightness is added. Then, the vehicle detection process ends.

Next, the vehicle position detection processing of FIG. 3 will be described in detail with reference to FIGS. 6 and 7.

FIG. 7 shows a flowchart of the vehicle position detecting process of FIG.

Regarding the vehicle position detection, first, S61
In, the first peak is detected for each added value of the horizontal luminance added in the addition processing, and the position Y ₁ that is the first peak is detected (FIG. 6). This first peak is the maximum value of the added values of the horizontal luminance added in the addition processing, and can be obtained by detecting this maximum value.

Then, the flow shifts to S62 in FIG. 7, where the second peak is detected for each added value of the horizontal luminances added in the addition processing, and the position Y _{2 which} is the second peak is detected. Detected (FIG. 6). The position of the second peak here is the peak of the added value in the Y position area separated by a predetermined distance from the position of the first peak, and is the position having the next largest value after the value of the first peak. Is.

After that, the processing shifts to S63 in FIG.
Position Y _{1 that} is the peak of Y and position Y ₂ that is the second peak
The position Y _{3 that} is the minimum value between and is detected (FIG. 6).

Then, the flow shifts to S64 of FIG. 7, where the position of the vehicle 3 is detected with the position Y ₃ having the minimum value as the head portion of the vehicle 3.

As described above, according to this embodiment, when it is determined that there is road reflection on the road 2, the horizontal luminance is added to each pixel of the processed image obtained by edge processing the image. The value obtained by adding the brightness in the lateral direction tends to be large (having the first and second peaks) at the position corresponding to the vehicle 3 and the reflection image 3a thereof. vice versa,
At the position corresponding to the boundary between the vehicle 3 and its reflection image 3a, this value tends to be relatively smaller than the value at the peak. By utilizing these tendencies, the boundary position between the vehicle 3 and the reflection image 3a can be accurately determined. Therefore, the position of the vehicle 3 can be accurately detected in the photographed image even in rainy weather.

[0033]

As described in detail above, according to the present invention, the position of a vehicle can be accurately detected in a photographed image even in rainy weather.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a vehicle detection device used for implementing a vehicle detection method according to the present embodiment.

FIG. 2 is a diagram showing a captured image captured by a CCD camera.

FIG. 3 is a flowchart showing image processing and vehicle detection processing in the vehicle detection method according to the present embodiment.

FIG. 4 is a diagram showing a cut-out image obtained by cut-out processing in the vehicle detection method according to the present embodiment.

FIG. 5 is a diagram showing a clipped image obtained by edge processing in the vehicle detection method according to the present embodiment.

FIG. 6 is a diagram showing each added value subjected to addition processing in the vehicle detection method according to the present embodiment.

FIG. 7 is a flowchart of vehicle position detection processing in FIG.

[Explanation of symbols]

1 ... Vehicle detection device, 2 ... Road, 3 ... Vehicle, 10 ... CCD
Camera, 20 ... Image processing device, 23 ... Image processing unit, 24
... Vehicle detection unit.

Claims (2)

[Claims] 1. A road surface determination step of determining whether or not there is road surface reflection of the road based on the brightness of pixels of an image of a vehicle traveling on the road, and it is determined that there is road surface reflection of the road. At this time, a vehicle detection method comprising: an addition step of adding brightness in the horizontal direction to each pixel of a processed image obtained by edge-processing the image; and a detection step of detecting the position of the vehicle based on a result of the addition. . 2. The detecting step detects the first peak and the second peak for each added value of the horizontal luminances added in the adding step, and then determines the position that becomes the first peak and the position where the first peak is reached. The vehicle detection method according to claim 1, wherein a position having a minimum value between the position having the second peak and the position having the minimum value is detected, and the position of the vehicle is detected with the position having the minimum value as the head portion of the vehicle.