JPH0371399A - Vehicle detecting method - Google Patents

Abstract

PURPOSE:To stably extract only the light source of a vehicle such as the headlight, the tail lamp, etc., that is, the features of the vehicle at night or in a dark place by attaching a polarizing element whose polarizing surface is set in the horizontal direction at the front of a TV camera and photographing a monitor subject area. CONSTITUTION:An image pickup means 1 catches a monitor subject area in the form of an image. A polarizing element 2 whose polarizing surface is set in the horizontal direction is set at the front of the means 1. An image storage means 3 stores the horizontally polarized image outputted from the means 1 in the form of the digital image data. The image data is sent to an undesired irradiating area deleting means 4. The means 4 applies a threshold value process to the received image data to delete the undesired irradiating areas and obtains the image data on only the headlight and tail lamp. The obtained image data is sent to a vehicle detection means 5 to decide a vehicle. Thus it is possible to stably extract only the vehicle light source like the headlight and the tail lamp, etc., serving as the features of a vehicle at night or in a dark place. Then the vehicle detecting accuracy is improved.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a vehicle detection method for detecting a vehicle existing in a predetermined monitoring target area such as a road or a parking lot at night or in a dark place.

[Conventional technology]

As a method for detecting vehicles traveling at night, a method has been proposed in which vehicles are detected by extracting vehicle light sources such as vehicle headlights and tail lamps from images taken with a television camera. For example, in JP-A-63-184898, a vehicle driving at night is captured as a multivalued image using a television camera or the like, and the brightness and darkness of each pixel are projected and summed in the vertical direction, and the maximum value of the addition result is An apparatus has been proposed that detects a vehicle by binarizing a vertically added image using a threshold determined based on the above-mentioned information and applying a predetermined vehicle determination condition to the binarized image.

[Problem to be solved by the invention]

However, in the vehicle detection method of the conventional device described above, when two vehicles are running in front of each other as shown in FIG. If a certain area is irradiated (the part shown by the hunting in Figure 10<al), the vertical density addition result will be as shown in Figure 1O (bl), and the irradiation brightness from the headlight will be compared to the brightness of the tail lamp. Because of the high value, the area illuminated by headlights is detected as the maximum value of the added density. Therefore, the threshold value is set based on this maximum value, so when threshold processing is performed using this value, , as shown in FIG. 10(e), the portion illuminated by the headlight of the following vehicle is extracted, and the information on the original tail lamp portion cannot be extracted, which may lead to incorrect determination.

In addition, in the case of rainy weather, if there is a puddle on the road surface as shown in Figure 11 (al), the reflected light from tail lamps, night lights, etc. from this puddle will also be captured as an image. The density addition result is as shown in FIG. 11(b), and even in such a case, information on the tail lamp portion cannot be extracted as shown in FIG. 11(c).

As mentioned above, with conventional methods, the presence of reflected light with a brightness equal to or higher than that of headlights or tail lamps can cause false detection, and it is not always possible to stably detect vehicle light sources such as headlights or tail lamps. The problem was that I couldn't do it.

The present invention was developed based on the above circumstances, and it removes so-called unnecessary reflection areas such as the vehicle body surface or the road surface with puddles that cause reflected light from the image, and identifies the characteristics of the vehicle at night or in a dark place. The purpose of the present invention is to provide a vehicle detection method that can stably extract only a vehicle light source such as a certain headlight or tail lamp.

[Means to solve the problem]

In the first vehicle detection method of the present invention, a polarizing element with a polarization plane set in the horizontal direction is attached to the front of a television camera to image a monitoring target area. Then, unnecessary irradiation areas other than vehicle light sources such as headlights and tail lamps are removed by performing predetermined threshold processing on the density values of the obtained horizontally polarized image.

In addition, in the second vehicle detection method of the present invention, in order to more reliably remove the unnecessary irradiation area, a polarizing element whose polarization plane can be changed horizontally and vertically is attached to the front of the television camera, and this polarizing element is The plane of polarization is alternately switched between horizontal and vertical directions, and the area to be monitored is photographed in synchronization with the switching of the plane of polarization. By calculating the density ratio of each corresponding pixel of the horizontal polarization image and vertical polarization image obtained in this way, and applying a predetermined threshold value processing to the image made up of the calculated density ratio, headlight The removal of unnecessary illumination areas other than vehicle light sources such as lights and tail lamps has been made more stable.

[Function] Before explaining the principle of the vehicle detection method of the present invention, the refraction and reflection of light and polarization, which are the basis of the unnecessary irradiation area removal process in the present invention, will be described.

Generally, the plane of vibration of natural light is random, but in order to make it easier to understand, in the following we will use vertically polarized light (S-polarized light) that vibrates perpendicular to the plane of incidence as shown in Figure 3.
2 of horizontally polarized light (P polarized light) that vibrates parallel to the plane of incidence
Pick up only one light.

Now, in Fig. 4 (al, l1yL, lf are the light intensities of the incident light 9, reflected light, and refracted light, M,, M,.

If M is the area of the incident light, 2 the area of the reflected light, and the area of the refracted light, then the incident light 1i, g the reflected light I,..., the refracted light ■.

The respective energies Wi pw and pWf can be expressed by the following equations.

Further, the energy reflectance R of the object i[ on which light is incident is defined as the following equation.

Here, when calculating the energy reflectances R, , R, of horizontally polarized light (P polarized light) and vertically polarized light (S polarized light) for any incident angle φ from Fresnel's formula, the following equations are obtained.

Therefore, using the relationship in equation (4), φ, γ
are the incident angle and refraction angle shown in Fig. 4).

From this equation (3), when (φ+γ) becomes π/2, R,
The denominator of tan'' (φ+γ〉 becomes infinite, so the energy reflectance R of horizontally polarized light (P-polarized light) becomes 0. In other words, when (φ+γ) is π/2, all horizontally polarized light components are transmitted. However, the reflected light reflected from substance A is only the vertically polarized (S-polarized) component. Expressing this in the formula, π φ+T= □ ~----・-・
(4) becomes. From 5nell's law, the relationship with the refractive index n1n2 in FIG. 4(b) is n=sin φ n.

-tan r °-゛----
(6)2 If the relationship of equation (6) holds true, all the horizontally polarized light components will pass through the material A, and the reflected light from the material A will be only the vertically polarized light.

The above content will be explained as follows using the actual photographed image shown in FIG.

That is, when formula (6) is satisfied, all the horizontally polarized light components of the incident light are transmitted through the puddle and absorbed by the road surface below, and the energy of the reflected light is attenuated accordingly. Therefore, the brightness and darkness values of the reflected light from the puddle are higher in the horizontally polarized image (Fig. 5(a)) than in the vertically polarized image (Fig. 5(b)).
)> is that much smaller (and the unnecessary irradiation area becomes darker as an image).

This phenomenon also applies to reflections on the vehicle body surface, although there are differences in degree.

In other words, in a horizontally polarized image, the difference in brightness and darkness between direct vehicle light sources such as headlights and tail lamps and reflected light from other unnecessary irradiation areas becomes larger, so horizontally polarized light with a larger density difference becomes By performing threshold processing using an image, it is possible to remove unnecessary irradiation areas and obtain an image consisting only of vehicle light sources such as headlights and tail lamps.

The present invention has been made based on the above theory, and the first vehicle detection method captures the monitoring target area as an image using the imaging means 1, as the principle is shown in FIG. A polarizing element 2 whose plane of polarization is set in the horizontal direction is arranged in front of the imaging means 1. The image storage means 3 stores the horizontally polarized image outputted from the imaging means 1 as digital image data, and sends the image data to the unnecessary irradiation area removal means 4. The unnecessary irradiation area removing means 4 removes unnecessary irradiation areas by applying threshold processing to the sent image data, and obtains image data of only the headlights and tail lamps. Then, the obtained image data is sent to the vehicle detection means 5,
Perform vehicle judgment.

Furthermore, the second vehicle detection method of the present invention also utilizes a vertically polarized image in order to more stably remove the unnecessary irradiation area.

That is, when comparing a horizontally polarized image and a vertically polarized image, there is almost no difference in the brightness and darkness density values of vehicle light sources such as headlights and tail lamps. The reason for this is that the vehicle light source emits direct light, and since the plane of vibration of the light is random, there is almost no effect on the polarization component.

On the other hand, as described above, there is a large difference in image density values of unnecessary irradiation areas other than vehicle light sources such as headlights and tail lamps between the horizontally polarized image and the vertically polarized image. - As an example, the reflectance of water is shown in FIG. In the figure, Rp is the reflectance for horizontally polarized light, and Rs is the reflectance for vertically polarized light. As is clear from this figure, the reflectance of horizontally polarized light is extremely lower than that of vertically polarized light. That is, in the horizontally polarized image, the brightness and darkness values of the unnecessary irradiation area are extremely small compared to the vertically polarized image. Therefore, when calculating the density ratio between these two images of vertically and horizontally polarized light, the density ratio of the vehicle light source parts such as headlights and tail lamps that are directly illuminated does not change much, but the density ratio of unnecessary irradiation areas becomes a large value. Become. Therefore, by setting a threshold value at the boundary between the vehicle light source and the unnecessary irradiation area in an image with this density ratio, it becomes possible to more reliably remove the unnecessary irradiation area from the screen.

The second vehicle detection method of the present invention utilizes the above theory, and as shown in FIG. 2, the monitoring target area is captured as an image by the imaging means 11.

At this time, the polarizing element 12 arranged in front of the imaging means 11
The polarization angle is alternately switched between the horizontal direction and the vertical direction by the polarization angle control means 16. Therefore, horizontally polarized images and vertically polarized images are alternately input to the imaging means 11, and each image is converted into a video signal that provides density in the imaging means 11.

The image storage means 13 stores the horizontally polarized image and the vertically polarized image outputted from the imaging means 11 as digital image data, and sends this image data to the unnecessary irradiation area removing means 14. The unnecessary irradiation area removing means 14 calculates a density ratio for each corresponding pixel using the density data of the horizontally polarized image and the vertically polarized image, and performs a predetermined threshold processing on the image formed by this density ratio. Unnecessary irradiation areas are removed and image data consisting only of vehicle light sources such as headlights and tail lamps is obtained. This image data is then sent to the vehicle detection means 15 for vehicle determination.

[Embodiment 1] Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.

FIG. 7 shows an example of a vehicle detection device configured by applying the first vehicle detection method of the present invention. In the figure, 21 is a television camera such as an ITV camera as an imaging means, and 22 is the front side of the television camera 21. 23 is a low-pass filter, 24 is an AD converter, 25 is an image memory that can store image data of one or more screens of input images, 26 is an output memory that stores calculation results, and 27 is a vehicle determination device. A microprocessor that performs processing.

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

The plane of polarization of the linear polarizer 22 is set in the horizontal direction, and the area to be monitored is photographed with the television camera 21. This photographed video signal is sent to an AD converter 24 through a low-pass filter 23. Then, the AD converter 24 samples the video signal of this horizontally polarized image, converts it into multi-gradation digital density data a, and sends it to the image memory 25.

Note that the low-pass filter 23 is inserted for the purpose of removing high-frequency noise components. - In general, noise generated from amplifiers inside imaging devices such as television cameras exists almost uniformly from high frequency components to low frequency components, so by removing this high frequency component with the low-pass filter 23, random noise In addition to reducing components, this also prevents aliasing noise from occurring during sampling.

When the digital density data for one screen of the horizontally polarized image has been captured as described above, the digital density data stored in the image memory 25 is sent to the output memory 26 via the data bus b. microprocessor 27
When it receives the computation end signal C, it reads the image data from the output memory 26 through the data bus d, performs threshold processing using the threshold value set based on this image data, removes unnecessary irradiation areas, and Obtain image data consisting only of light sources such as lights and tail lamps. Then, the vehicle determination described below is executed based on this image data, and when the vehicle determination process is completed, the microprocessor 27 outputs a process end signal e to the AD converter 24 and repeats the above-mentioned processes again.

FIG. 9 is a flowchart of the vehicle determination process by the microprocessor 27 described above. After the unnecessary irradiation area removal process described above, the microprocessor 27 specifies a detection area for vehicle determination (step (1)), and performs vertical projection summation of the density values of each pixel existing within this specified area. (Step [2]).

Then, in this addition result, if there is a positive polarity of extremely high brightness, it is determined that "vehicle is present" and step (3)
)(4)), and if it does not exist, it is determined that there is no vehicle.In steps (3) and (5)), a vehicle is detected.

By arranging the linear polarizer horizontally in front of the television camera as described above, it is possible to remove unnecessary irradiation areas other than vehicle light sources such as headlights and tail lamps, and perform stable vehicle determination.

[Example 2] Next, a second embodiment of the present invention will be described.

FIG. 8 shows an example of a vehicle detection device constructed by applying the second vehicle detection method of the present invention. 33 is a low-pass filter, 34 is an AD converter, 35.36 is an image memory each capable of storing image data of one or more screens of human images, 37.38 is a logarithm calculator, and 39 is a difference calculator. Arithmetic unit, 40
41 is a microprocessor that performs vehicle determination processing; 42 is a control unit that performs variable control of the polarization angle of the linear polarizer 32 and sampling control for capturing image data; 43 is a sampling timer; be.

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

The television camera 31 captures an image of the area to be monitored. The control unit 42 outputs a polarization angle control signal a, rotates the linear polarizer 32 to a vertical polarization position, and outputs an image capture signal S to the AD converter 34 in synchronization with this.

When the linear polarizer 32 is set to the vertical polarization position, the television camera 31 outputs a video signal of a vertically polarized image,
This video signal is sent to an AD converter 34 through a low-pass filter 33. Then, the AD converter 34 samples the video signal of this vertically polarized image, converts it into multi-gradation digital density data C, and sends it to the image memories 35 and 36.

Note that the low-pass filter 33 is inserted for the purpose of removing high-frequency noise components, as in [Embodiment 1].

Next, the control unit 42 outputs the image memory selection signal d,
One of the two image memories 35 is selected, and digital density data for one screen of the vertically polarized image is stored in the image memory 35.

When the capture of one screen worth of digital density data of the vertically polarized image into the image memory 35 is completed, the control section 42 outputs a timer start signal k and starts the sampling timer 43. When the sampling timer 43 times up after a certain period of time has elapsed, it outputs a time-up signal e to the control section 42.

After receiving the time-up signal e, the control unit 42 outputs the polarization angle control signal a again, rotates the linear polarizer 32 to change its polarization plane from the vertical polarization position to the horizontal polarization position, and further synchronizes with this. The image capture signal is output to the AD converter 34.

When the linear polarizer 32 is set to the horizontal polarization position, the television camera 31 outputs a video signal of a horizontally polarized image,
The signal is sent to the AD converter 34 through the low-pass filter 33. Then, the AD converter 34 samples the video signal of this horizontally polarized image, converts it into multi-gradation digital density data C, and sends it to the image memories 35 and 36.

The control unit 42 outputs an image memory selection signal to select the image memory 36, and stores digital density data for one screen of the horizontal polarization image in the image memory 36.

When the digital density data for one screen of the vertically polarized image and the horizontally polarized image has been captured in the above manner, the digital density data stored in the image memory 35.36 is logarithmically operated by the logarithm calculator 37.38. , data buses g and h to the difference calculator 39.

The difference calculator 39 calculates the difference for each corresponding pixel of the vertically polarized image and the horizontally polarized image sent from the logarithm calculators 37 and 38, and outputs the logarithmic difference calculation value l to the memory 4.
Output to 0 and store. In this way, by logarithmically transforming the density data of the two images and then performing a difference calculation,
This means that the calculation process equivalent to division, that is, the same calculation as calculating the density ratio for each corresponding pixel of the vertically polarized image and the horizontally polarized image is performed.

When the logarithmic difference calculation is completed for one screen of the vertically polarized image and the horizontally polarized image, the difference calculator 39 outputs a calculation end signal i to the microprocessor 41.

When the microprocessor 4* receives the computation end signal i, it reads the logarithmic difference computation data from the output memory 4° via the data bus j, and performs a predetermined threshold processing based on the logarithmic difference computation data.

As an example of this threshold processing, for example, if the concentration ratio is smaller than the set threshold value, "O", and if it is larger, "l" °
And it is sufficient. As a result, image data from which unnecessary irradiation areas have been removed can be obtained. For this image data [Example 1
] Similarly, the vehicle determination process shown in FIG. 9 is executed. When the vehicle determination process is completed, the microprocessor 41 outputs a process end signal m to the control section 42, returns the state of the control section 42 to the initial state, and repeats the above-described processes again.

Generally, when the imaging conditions change, the image density value also changes, but by calculating the density ratio of the horizontally polarized image and the vertically polarized image as described above, the threshold value can be set without considering changes in the density value. Therefore, unnecessary irradiation areas can be removed more stably.

〔Effect of the invention〕

As is clear from the above, when using the vehicle detection method of the present invention, image information of unnecessary irradiation areas other than vehicle light sources such as headlights and tail lamps can be reliably extracted from the image, which was a major drawback of the conventional method. This makes it possible to stably extract only vehicle light sources such as headlights and tail lamps, which are characteristic of vehicles at night or in dark places, and improve vehicle detection accuracy.

[Brief explanation of drawings]

Fig. 1 is a principle diagram of the first vehicle detection method of the present invention, Fig. 2 is a principle diagram of the second vehicle detection method of the present invention, Fig. 3 is an explanatory diagram of horizontal polarization and vertical polarization, and Fig. 4 is an explanatory diagram of refraction and reflection of light, FIG. 5 is a diagram showing an example of a photographed image, FIG. 6 is a diagram of measuring the reflectance of water, and FIG. 7 is a block diagram of the first embodiment of the present invention. FIG. 8 is a block diagram of a second embodiment of the present invention, FIG. 9 is a flow chart of vehicle determination in the above embodiment, and FIGS. 10 and 11 are explanatory diagrams of vehicle erroneous detection due to unnecessary irradiation.

Claims (2)

[Claims] (1) A polarizing element with the plane of polarization set horizontally is placed in front of an imaging device such as a television camera, and a predetermined threshold value process is applied to the horizontally polarized image obtained by the imaging device. A vehicle detection method characterized by extracting an image consisting only of a vehicle light source such as a tail lamp, and determining a vehicle based on the extracted image. (2) A polarizing element is placed in front of an imaging means such as a television camera, and its plane of polarization is variably controlled in the horizontal and vertical directions, and the corresponding pixel positions of the horizontally polarized image and the vertically polarized image obtained by the imaging means are By calculating the density ratio and performing predetermined threshold processing on the image formed by this density ratio, an image consisting only of vehicle light sources such as headlights and tail lamps is extracted, and a vehicle is determined based on the extracted image. A vehicle detection method characterized by: