JP4936045B2 - Vehicle color discrimination device, method and program - Google Patents

Description

The present invention relates to a vehicle body color discriminating apparatus , method and program capable of discriminating color information of a vehicle by image processing.

Detailed traffic information provision is required for smooth traffic investigation support. By providing detailed vehicle type information, it is possible to track a specific vehicle.
In the conventional method of identifying the vehicle by the license plate, there is a problem that the vehicle cannot be reliably identified by itself because the license plate is dirty or bent.
Therefore, Patent Document 1 proposes a method for discriminating the paint color of a vehicle by image processing. In this method, the presence / absence of a vehicle is determined from the density distribution of a photographed image, and if a vehicle exists, a specific area of the vehicle (specifically, a portion where the density distribution is uniform) based on the density distribution. Is extracted, and the color of the specific area is set as the vehicle body color.

Moreover, in patent document 2, a vehicle color is discriminate | determined from the wavelength intensity distribution of the wavelength spectrum of the reflected light which irradiated the white light to the vehicle and received with the light-receiving part. Alternatively, the reflected wave is photographed with a color camera, and the average RGB value is used as the body color.
JP 2000-222673 A JP 2000-334665 A

When discriminating the vehicle body color, it is a big point which part of the vehicle the discrimination target is. Since the camera shoots as if looking down the road from above the road, it is desirable for a general vehicle to target a horizontal surface such as a roof or a hood.
However, if the average value of RGB is used to discriminate the body color as in Patent Document 2, non-colored areas such as the front grille and the windshield are also targeted, causing erroneous determination.

When a portion having a uniform density distribution (covered) as in Patent Document 1, the windshield is covered, so the colors of the windshield and the front grill are also discriminated, which is also the cause of erroneous determination. It becomes.
Accordingly, an object of the present invention is to provide a vehicle body color discrimination device , method, and program capable of measuring a vehicle and discriminating the vehicle body color for an area determined to be a horizontal plane.

The vehicle body color discriminating apparatus of the present invention includes a photographing means for photographing a vehicle traveling on a road, a horizontal plane extracting means for extracting a horizontal plane of a vehicle body of the vehicle traveling on the road, and a color of the horizontal plane extracted by the horizontal plane extracting means. A horizontal plane color discriminating unit that discriminates based on a photographed image of the photographing unit, wherein the horizontal plane extracting unit detects a line portion of the vehicle body based on a photographed image by the photographing unit, and detects the detected line portion. calculating a height substantially the plane sandwiched by two line sections of the same height as the horizontal plane, characterized in that to determine the color of the vehicle body based on the color of the horizontal plane is determined by the horizontal color discriminating means .

Here, the “horizontal plane” is not only a strict horizontal plane but also a concept including a surface having a gentle convex surface, such as a vehicle roof or a bonnet, and a surface inclined by a slight angle.
In the present invention, the horizontal plane is extracted to discriminate the vehicle body color. For example, a roof or a bonnet. If these horizontal planes are targeted, there is an advantage that the vehicle body color can be clearly distinguished.

Based on the shooting image, calculates the height and length of the line section and substantially a same height and horizontal plane surface to be sandwiched between the two line sections of approximately the same length, more reliably horizontal plane particular it can.

In some cases, a plurality of line portions having different heights are extracted from one vehicle body. If there are a plurality of line portions having different heights, the horizontal plane extracting means may extract a horizontal plane preferentially from a pair of line portions having a high height. This is because, from experience, the higher the horizontal plane, the larger the area, and thus the color can be easily distinguished.

If there are a plurality of line portions having different distances between the line portions along the vehicle traveling direction, the line portions may be extracted from a horizontal plane having a long distance between the two line portions. This is because, from experience, it is easier to discriminate the color as the horizontal surface of the vehicle body increases.
When the extracted horizontal plane exists between the line portions having the longest distance, it is preferable not to extract the horizontal plane based on the two line portions. This is because when a horizontal plane has already been extracted in a portion surrounded by two line portions, there can be no horizontal plane surrounding the horizontal plane.

If there are a plurality of horizontal planes having different lengths in the vehicle transverse direction, the horizontal plane may be extracted preferentially from a pair of long line portions. This is because, from experience, a horizontal plane with a wider width has a larger area, which makes it easier to distinguish colors.
Further, it is preferable to determine whether the screen is backlit or follow light as viewed from the camera, and to change the threshold value for determining the brightness of the horizontal plane according to the follow light / back light. Since the brightness of the screen looks different depending on the forward light / backlight, the accuracy of color discrimination is improved by performing such correction.

In addition, a horizontal plane with a small area may be excluded from the horizontal plane color discrimination target. This is because a horizontal plane having a small area may be extracted by mistake due to a detection error.
Depending on how light is reflected / absorbed, the color may vary depending on the level of the horizontal plane. In this case, the color having the largest number of pixels among the pixels constituting the horizontal plane may be determined as the vehicle body color.
A condition may be that the number of pixels of the color is equal to or greater than a certain percentage of the number of pixels in the entire horizontal plane. If the color varies widely depending on the part of the horizontal plane, colors exceeding a certain ratio may not be discriminated, but in this case, it is excluded from the color discrimination target.

Further, when the determined colors are different for a plurality of horizontal planes, the color of the horizontal plane having the largest area may be set as the vehicle body color. This is because the color of the horizontal plane having the largest area is considered to represent the color of the vehicle most accurately.
Further, when the determined colors are different for a plurality of horizontal planes, a plurality of vehicle body colors may be output for each of the plurality of horizontal plane colors. This is because some vehicles have different colors for a plurality of horizontal planes. For example, the color of the roof and bonnet are different.

  The vehicle body color discrimination method and program of the present invention are the method and program according to the same invention as the invention of the vehicle body color discrimination device.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 shows an installation view of a vehicle body color discrimination device of the present invention.
The camera 2 is installed at a predetermined height so as to overlook the road surface above the road. The visual field range 5 of the camera 2 is on one lane of the road.
The captured image of the camera 2 is input to the measuring device 4 through the video cable 3. The measuring device 4 has communication means and can provide the measurement result to a remote processing center.

In FIG. 2, the example of the picked-up image image | photographed with the camera 2 is shown.
The visual field range 5 of the camera 2 is a visual field (about 3 to 4 m in the left-right direction and about 5 to 6 m in the vertical direction) such as the hood and roof of the vehicle that can easily extract the horizontal plane of the vehicle.
FIG. 3 is a block diagram illustrating a configuration example of the measuring device 4.
The measuring device 4 captures the image of the camera 2, determines the presence or absence of a vehicle within the vehicle detection measurement range set on the screen of the camera 2, and detects a horizontal line when a vehicle is detected. Calculation processing such as identification of the vehicle body and discrimination of the vehicle body color.

The measurement device 4 includes an image input unit 41, an A / D conversion unit 42, an image memory 43, a calculation unit 44, and a communication unit 45.
The image input unit 41 provides an interface for capturing an image of the camera 2. The A / D converter 42 digitally converts the image signal. The image memory 43 temporarily stores the digitally converted image data. The calculation unit 44 executes calculation processes such as detection of a horizontal line, specification of a horizontal plane, and vehicle body color determination. The communication part 45 communicates with a traffic light, a display board, a traffic control center, etc., and conveys the information of a vehicle body color.

The camera 2 and the measuring device 4 may be integrated.
FIG. 4 is a flowchart showing a vehicle body color discrimination processing method performed by the measuring device 4. All or some of the functions of the respective units of the measuring device 4 are realized by the calculation unit 44 of the measuring device 4 executing a program recorded on a predetermined medium such as a CD-ROM or a hard disk.
The calculation unit 44 captures an image through the image input unit 41 and stores it in the image memory 43 (step S1). The image capture interval is, for example, 30 images per second.

Based on the image stored in the image memory 43, the calculation unit 44 detects whether or not a vehicle is present on the screen (step S2).
For example, a road surface image (hereinafter referred to as a background image) when a vehicle is not present is created in advance, a brightness difference from the captured image (hereinafter referred to as an input image) is calculated, and a change that exceeds a threshold value Is detected, the vehicle is present. If the change is equal to or less than the threshold value, the vehicle is absent (step S3).

Alternatively, the position of a portion of the road surface pattern where the brightness changes (hereinafter referred to as a line) is stored in advance, the line is also calculated for the input image, the difference in the sharpness of those lines is calculated, When the difference is detected, the vehicle can be present, and when the difference is less than or equal to the threshold, the vehicle can be absent.
If it is determined that there is a vehicle, the vehicle speed is detected (step S4). Known means can be used for vehicle speed detection. For example, an ultrasonic Doppler radar can be installed on the camera 2 side to measure the speed of the vehicle. In addition, it is possible to know the speed of the vehicle by installing two embedded vehicle detectors on the road and measuring the time when the vehicle passes through these vehicle detectors.

After the speed detection, horizontal line detection processing and horizontal plane extraction processing are performed (steps S5 to S7).
The horizontal line detection process (step S5) will be described in detail with reference to FIG.
First, for each pixel, the brightness difference between the previous time and the current time is calculated. If the difference is equal to or greater than a threshold, it is determined that there is a time difference change (step T1). This is for erasing the line of a stationary road surface such as a road marking.

Next, the presence / absence of a horizontal edge is calculated for a pixel with a time difference change (step T2). For example, as shown in FIG. 6, when the Sobel filter process is performed on nine pixels of the target pixel and its surrounding pixels, and the absolute value of the calculated value D is equal to or greater than the threshold value, the target pixel (*) Suppose there is a horizontal line.
The calculation formula is, for example, D = I11 × 1 + I12 × 2 + I13 × 1-I31 × 1-I32 × 2-I33 × 1
(Ijk represents the luminance of the pixel in the j-th row and the k-th column of the matrix).

Next, there is a time difference change, and thinning processing is performed on the pixels constituting the horizontal line (step T3). This is a process for removing adjacent lines and making it easier to detect horizontal lines. For example, a horizontal line image is thinned using Hilditch's thinning algorithm (see “Processing and Recognition of Images”, Yasuiin et al., Shosodo).
FIG. 7 is a diagram illustrating an example of a thinned horizontal line image.

Next, the number of horizontal lines is calculated in the horizontal direction of the screen, and a histogram is created as shown at the right end of FIG. 7 (step T4).
Based on this histogram, from the upper part to the lower part of the screen (or from the lower part to the upper part), a line whose number of horizontal lines is larger than the threshold value by comparison with the horizontal lines adjacent to the top and bottom of the horizontal line is detected (step T5). . This corresponds to a sharp peak portion surrounded by a circle in FIG. This line is called a “horizontal line”.

However, short horizontal lines (for example, 1 m or less) are excluded. This is because a short horizontal line is likely not a line belonging to the vehicle hood or roof.
When the horizontal line detection process is completed in this manner, the process returns to FIG. 4 to enter the horizontal line height calculation process in step S6. In this height calculation process, vehicle speed information is used.
Note that the horizontal line height calculation described below is an example of a method using vehicle speed information and a single monocular camera.

The horizontal line height calculation process will be described in detail below.
FIG. 8 is a graph with the position on the screen of the detected horizontal line on the vertical axis and the time on the horizontal axis. The case where three horizontal lines are shown is illustrated.
In FIG. 8, the trajectory of each horizontal line is an apparent movement on the screen, and is a projection of the position in real space at each time. As the vehicle travels, the horizontal line moves from the top to the bottom of the screen.

Here, a conversion formula between coordinates in the real space and the imaging plane is defined.
As shown in FIG. 9, the coordinate system in the real space is (x, y, z), and the coordinate system on the imaging surface of the camera 2 is (p, q).
In real space, the vertical direction is z, the direction in which the road extends is y, and the direction across the road is x.
The camera 2 is installed on an extended line of the road, and on the photographing plane thereof, the horizontal axis p is parallel to the road crossing direction x, and the vertical axis q is perpendicular to the horizontal axis p. It is assumed that the vertical axis q is inclined by an angle α corresponding to the depression angle with respect to the vertical direction z of the road.

Here, since it is assumed that the camera 2 is installed on an extension line of the road and the vehicle approaches the camera 2, the movement in the road crossing direction x in the real space is not considered. On the imaging plane, only the horizontal line position coordinate q is considered, and the coordinate p orthogonal thereto is ignored. Therefore, the coordinates of the vehicle in the real space are (y, z) and q on the photographing surface.
The following relational expressions (1) and (2) are established for the coordinate q (unit: number of pixels) on the imaging surface and the coordinate point (y, z) on the real space.

However, the ratio of the focal length f of the lens of the camera 2, the installation height H of the camera 2, the height Sq in the q-axis direction on the imaging surface, and the number Nq of pixels in the q-axis direction of the imaging surface corresponding to Sq ( The size of one pixel) is known and treated as a constant value.
z = H− (ftan α−ξ) y / (ξtan me {f) (1)
ξ = (Sq / Nq) q (2)
Here, ξ is the height of the horizontal line on the imaging surface. Since the moving speed v of the vehicle within the minute time (the time passing through the upper and lower range of the screen) has been detected, the y coordinate y (t) at each time can be calculated based on the elapsed time from the predetermined time.

The height z in the real space of the horizontal line is assumed to be a certain constant value.
Then, from equation (1), ξ (t) can be calculated as a function of time based on y (t). Using this ξ (t), q (t) can be obtained as a function of time from equation (2).
Therefore, the curve U can be drawn using q (t) as a function of time.
If the height z of the horizontal line in real space is used as a variable (parameter) and the height z is changed, a curve group depicting a plurality of q (t) can be formed.

FIG. 10 is a graph showing the curve group U of these q (t) with the height z of the horizontal line in real space as a variable.
In the figure, the height on the photographing surface at time t0 is unified to the same value qs for each curve. In the same figure, the height z has a relationship of z0>z1> z2.
It can be seen that the curve U0 corresponding to the height z0 passes through the imaging surface from the upper end to the lower end (referred to as “field of view of the camera 2” from the upper end to the lower end of the imaging surface) in a short time. This phenomenon corresponds to an empirical rule that a high object such as a vehicle roof crosses the field of view of the camera 2 in a relatively short time.

On the other hand, an object with a low height, such as a bumper of a vehicle, passes through the field of view of the camera 2 in a relatively long time even when traveling at the same speed.
In the graph of FIG. 10, the curve U2 corresponding to the height z2 passes through the imaging surface over a long time.
Therefore, assuming that the moving speed of the vehicle is known, the height of the object can be obtained by measuring the time T during which the horizontal line passes through the field of view of the camera 2.

Further, assuming that the moving speed of the vehicle is known, instead of the time T when the horizontal line passes through the field of view of the camera 2, the apparent speed at which the horizontal line passes through the field of view of the camera 2 (from the upper end to the lower end of the screen) The height of the object can also be obtained based on the distance (q0−q1) divided by the time T).
For example, in FIG. 10, examples of trajectories of horizontal lines are plotted, and it is assumed that it takes time T for these points to pass through the field of view of the camera 2. A curve U * passing through the field of view of the camera 2 over this time T can be specified. If the height corresponding to the curve U * is z *, that z * is the height of the horizontal line in real space.

In addition, instead of measuring the time T and the apparent speed, the trajectory of the horizontal line (that is, the shape of the curve U) can be known, so that the shape of this trajectory is used as each curve constituting the curve group U. The curve having the most similar inclination may be specified by fitting to. Then, the height corresponding to this curve is known.
For example, in FIG. 10, the trajectory of the photographed horizontal line is plotted every moment, and these points are assumed to fit the curve U *. If the height corresponding to the curve U * is z *, the z * is obtained as the height of the horizontal line in real space.

Since the shape of the curve group is different if the moving speed of the vehicle is different, a curve group must be prepared for each moving speed of the vehicle.
However, if a group of curves corresponding to one moving speed is provided, a group of curves of other moving speeds can be easily obtained simply by extending or reducing the time scale.
In the above description, as shown in FIG. 10, the time was measured on the graph and the coincidence of the curve shapes was examined. However, these processes are actually performed using the computer processing function. Needless to say. For example, in order to check the coincidence of the curve shapes, a known maximum likelihood estimation method such as a least square method may be used.

As described above, the height z of the horizontal line in the real space is known, and the distance y from the vehicle to the camera 2 can be calculated by the equation (1) using the position q on the screen every moment. Thus, the coordinates (y, z) in the real space of the horizontal line are known.
The horizontal line detection process is not limited to the method described above. The method described above is a method using one camera (monocular camera). However, for example, the height of an object in real space can be obtained using two cameras (compound-eye camera). .

  In this method, a vehicle is photographed using two cameras arranged vertically, and images from each camera are sequentially captured and three-dimensional measurement is performed using the principle of triangulation. This is a method for obtaining three-dimensional coordinates. Two cameras arranged vertically are installed in parallel, have the same focal length, and each imaging surface is located on the same plane. Therefore, the imaging position on each input image concerning a predetermined object point in the space is shifted only in the vertical direction. Based on this deviation, the three-dimensional coordinates of the object point in the real space can be obtained (see Japanese Patent Application Laid-Open No. 11-259792 [0042]-[0051]).

Thus, when using a monocular camera or using a compound eye camera, there is an advantage that an increase in cost can be suppressed if an existing camera can be used.
Next, the horizontal plane extraction process in step S7 of FIG. 4 is performed.
This process is a process in which a pair having a close height or a pair having a close height and length is selected from the horizontal lines, and a surface sandwiched between them is set as a horizontal plane.

  FIG. 11 is a diagram illustrating a state in which five horizontal lines are detected from a captured image of the vehicle body. Horizontal lines are indicated by bold lines in the figure. The lines having the same height are the first and second lines from the top, and the third and fourth lines from the top. In this way, two pairs are extracted. Note that the same height means that the two heights are h1 and h2, for example, the absolute value | h1−h2 | of the height difference is equal to or less than the threshold value, or the difference between the height ratio and 1 The case where the absolute value | (h1 / h2) −1 | The threshold value may be selected such that a surface that is slightly inclined such as a roof or a bonnet of a vehicle is extracted and a surface that is largely inclined such as a windshield or rear glass is not extracted.

Even if the extracted pairs are the same height, if the lengths are different, there is a high possibility that they do not belong to the same horizontal plane.
Note that the same length means that the two lengths are x1 and x2, for example, the absolute value of the length difference | x1-x2 | is equal to or less than the threshold value, or the difference between the length ratio and 1 This is the case where the absolute value | (x1 / x2) -1 |

If there are a plurality of pairs having different heights, a horizontal plane may be extracted preferentially from a pair of horizontal lines having a high height. This is because in the case of a car body, the tall roof has a wider horizontal surface and is likely to be paired.
For example, as shown in FIG. 12, the roof of the vehicle is extracted by a pair of first and second lines from the top. The bonnet formed by the first and second pair of lines from the top is also extracted. Since the height of the vehicle roof is higher, the vehicle roof is preferentially extracted.

If there are a plurality of horizontal lines with different distances along the vehicle traveling direction (y direction), it is preferable to extract them from a horizontal plane having a long distance between the two lines. If there are a plurality of horizontal lines having different lengths in the vehicle transverse direction (x direction), the horizontal plane may be extracted preferentially from a pair of horizontal lines having a long length. This is because it would be desirable to extract as much of the horizontal plane as possible.
However, even if a pair is made, it excludes when there is already a horizontal plane with a different height between them. This is because if there are already horizontal planes with different heights, there cannot be a horizontal plane that includes them.

When the horizontal plane extraction process ends, the vehicle detection count is incremented by 1 (FIG. 4; step S8), and the process returns to step S1.
Thereafter, the processing from step S2 onward is repeated based on the next captured image.
If it is determined in step S3 that there are no vehicles in the image, the process proceeds to step S9, where the number of vehicle detections so far is compared with a threshold value n.

The threshold value n of the number of vehicle detections is set to about “3”, for example. This is to stop the following body color output processing because it is doubtful that the vehicle is really a vehicle if the continuous vehicle detection can be performed only three times or less.
If the number of vehicle detections exceeds the threshold value n, the process moves to step S10, and step body color determination processing is performed.

FIG. 13 is a detailed flowchart for explaining the vehicle body color determination process.
The horizontal planes extracted so far are processed in order (step U1). When a horizontal plane is extracted from each of a plurality of images, even a horizontal plane of the same part of the vehicle body is treated as a separate horizontal plane.
In order to exclude a horizontal surface with a small area, it is compared with a reference value (for example, 1 m ² ) (step U2). This is because a small horizontal plane cannot sufficiently distinguish colors.

Next, the pixels included in the horizontal plane are sequentially processed (step U3).
The color of one pixel is determined (step U4). In order to determine the color of a pixel, RGB is converted into HSI (Hue, Saturation, Luminance (Value)). However, R, G, B, S, and I are values from 0 to 1, and H is an angle of 0 ° ≦ H <360 °.
Next, it classifies into each color according to the HIS value. Here, a case where classification is made into 11 colors of red / orange / yellowish green / green / cyan / blue / purple / magenta / black / gray / white is given as an example.

First, it is classified into a chromatic color or an achromatic color according to S (saturation) of each pixel. For example, a chromatic color is set when S ≧ 0.2, and an achromatic color is set when S <0.2.
If it is determined as an achromatic color, it is classified into black / gray / white according to the following criteria according to I (luminance). Black: 0.0 ≦ I ≦ 0.2, Gray: 0.2 <I ≦ 0.8, White: 0.8 <I ≦ 1.0.
However, the threshold value (0.2, 0.8) for classification as black / gray / white may be changed according to the forward light / backlight. The position of the sun is determined from the position and direction of the camera and the current date and time, and it is determined from the camera whether it is backlit or follow light. During backlighting, the horizontal plane appears brighter, so the threshold is raised.

In the case of a chromatic color, it is classified according to H whether it corresponds to red / orange / yellowish green / green / cyan / blue / purple / magenta.
Next, the color of the horizontal plane is determined (step U6). When both of the following conditions (1) and (2) are satisfied, the color of the horizontal plane is determined. When there is no corresponding color, the color of the horizontal plane cannot be determined.
(1) The number of pixels of the color is the largest. (Maximum frequency)
(2) The number of pixels of the color is not less than a certain ratio (for example, 40%) of the number of pixels in the entire horizontal plane.

The color of the vehicle body is determined based on one of the following criteria (step U9).
(a) The color of the horizontal plane with the largest area is the vehicle body color. (Output only one color)
(b) Output colors for each horizontal plane such as bonnet and roof. (All colors are output)
In this way, the vehicle body color is determined.
When the vehicle body color is determined, information on the determined vehicle body color is output via the communication unit 45 in step S11 of FIG.

Although the embodiments of the present invention have been described above, the embodiments of the present invention are not limited to the above-described embodiments. For example, the horizontal plane can be directly extracted by detecting the height of each pixel by laser scanning.
This laser scanning method will be described with reference to FIGS.
FIG. 14 is a coordinate diagram showing the relationship between the coordinate system (x, y, z) in the real space and the coordinate system (p, q) on the imaging surface of the camera 2, and is substantially the same as FIG. It is the same figure.

The difference from FIG. 9 is that in this embodiment, a laser range finder 6 is attached to the camera 2, and the reflected wave of the object on the road is detected by laser irradiation of the laser range finder 6. The distance d to the object can be calculated.
Therefore, the coordinates p and q on the photographing screen of the object and the distance d to the object are known. If the coordinates p and q and the distance d are known, the height h of the object from the ground as shown in FIG. 14 can be obtained using camera parameters (camera position and direction: known).

It is assumed that the laser range finder 6 can scan within the photographing range of the camera. As shown in FIG. 15, the laser distance measuring points are distributed in a grid pattern on the imaging range of the camera.
Accordingly, the height h of the object from the ground can be plotted on the photographing range of the camera.
The horizontal plane can be extracted by collecting points having the same height h. This horizontal plane is shown by lane hatching in FIG.

If the horizontal plane is extracted, the vehicle body color can be determined using the vehicle body color determination process described with reference to FIG.
The advantage of this laser scanning method is that the horizontal plane is detected with higher accuracy than when the horizontal plane is extracted using only the camera, and the horizontal plane can be detected accurately. In addition, you may use together this laser scanning method and the method of extracting a horizontal surface using the camera mentioned above.

It is a block diagram of a vehicle body color discrimination device. It is a figure which shows the example of a picked-up image of a camera. It is a block diagram which shows the structural example of a measuring device. It is a flowchart for demonstrating the vehicle color discrimination method. It is a detailed flowchart for demonstrating a horizontal line detection process. It is a figure which shows the mask for a detection of a horizontal line. It is a figure which shows the thinned horizontal line image and the horizontal line number histogram. It is a locus diagram on the time space of each horizontal line position reflected on the screen. It is a coordinate diagram which defines the coordinate on real space and an imaging surface. It is a graph which shows the relationship q (t) between each horizontal line position q reflected on the screen and time t, using the height z of the horizontal line in real space as a variable. It is a figure which shows the horizontal line in the image | photographed vehicle body. It is a figure which shows the horizontal surface extracted combining the horizontal line position. It is a detailed flowchart which shows a vehicle body color discrimination | determination process. 3 is a coordinate diagram illustrating a relationship between a coordinate system in a real space and a coordinate system on a photographing surface of the camera 2. FIG. It is a distribution map of the laser ranging point on the imaging range of the camera.

Explanation of symbols

2 Camera 3 Video cable 4 Measuring device 6 Laser rangefinder 41 Image input unit 42 A / D conversion unit 43 Image memory 44 Calculation unit 45 Communication unit

Claims (15)

Photographing means for photographing a vehicle traveling on a road;
A horizontal plane extracting means for extracting a horizontal plane of a vehicle body of a vehicle traveling on a road;
A horizontal plane color discriminating unit for discriminating a color of a horizontal plane extracted by the horizontal plane extracting unit based on a photographed image of the photographing unit;
The horizontal plane extracting means detects a line portion of the vehicle body based on a photographed image by the photographing means , calculates a height of the detected line portion, and a surface sandwiched between two line portions having substantially the same height. A horizontal plane ,
Body color discriminating apparatus characterized by determining the color of the vehicle body based on the color of the horizontal plane is determined by the horizontal color discriminating means. Photographing means for photographing a vehicle traveling on a road;
A horizontal plane extracting means for extracting a horizontal plane of a vehicle body of a vehicle traveling on a road;
A horizontal plane color discriminating unit for discriminating a color of a horizontal plane extracted by the horizontal plane extracting unit based on a photographed image of the photographing unit;
The horizontal plane extracting means detects a line portion of the vehicle body based on a photographed image by the photographing means, calculates the height and length of the line portion, and has two of the same height and substantially the same length. The surface sandwiched between the line parts is a horizontal plane,
A vehicle body color determination device that determines the color of a vehicle body based on the color of the horizontal surface determined by the horizontal surface color determination means . The vehicle body color discriminating apparatus according to claim 1 or 2 , wherein the horizontal plane extracting means preferentially extracts a horizontal plane from a pair of high line portions if there are a plurality of line portions having different heights. The horizontal plane extracting means, if a plurality are different line portion of the distance between the line portions along the vehicle travel direction, one of claims 1 to continue to extract from a long horizontal a distance between the two line sections of claims 3 The vehicle body color discrimination device according to claim 1. The horizontal plane is not extracted based on the line portion having the longest distance between the two line portions when an extracted horizontal plane exists between the line portions having the longest distance between the two line portions. 4. The vehicle body color discrimination device according to 4 . 6. The horizontal plane extracting unit according to any one of claims 1 to 5 , wherein when there are a plurality of horizontal planes having different lengths in the vehicle transverse direction, the horizontal plane is extracted preferentially from a pair of long line portions. The vehicle body color discriminating device described in 1. The horizontal color determining means determines whether the backlight or front light as viewed from the camera, in accordance with the order light / backlight, any of claims 1 to 5 for varying the threshold to determine the brightness of the horizontal plane The vehicle body color discrimination device according to claim 1. The horizontal plane color discriminating means includes, among the pixels constituting the horizontal plane,
The vehicle body color discrimination device according to any one of claims 1 to 7 , wherein a color having the largest number of pixels of the color is discriminated from a horizontal plane color. The horizontal plane color discriminating means includes, among the pixels constituting the horizontal plane,
The vehicle body color discriminating apparatus according to claim 8 , wherein the color is discriminated on the condition that the number of pixels of the color is equal to or greater than a certain ratio of the number of pixels of the entire horizontal plane. The vehicle body color discriminating device according to any one of claims 1 to 9 , wherein a color of a horizontal plane having the largest area is a vehicle body color. The vehicle body color discriminating device according to any one of claims 1 to 9 , wherein a plurality of vehicle body colors are output for each of a plurality of horizontal plane colors. A surface sandwiched between two line portions having substantially the same height by detecting line portions of the vehicle body of the vehicle based on a photographed image of a vehicle traveling on a road, calculating the heights of the detected plurality of line portions. As a horizontal plane of the vehicle body of the vehicle,
Discriminate the color of the extracted horizontal plane based on the captured image,
A vehicle body color determination method, wherein the vehicle body color is determined based on the determined horizontal plane color. A program executed by a computer,
Capture an image of a vehicle traveling on the road,
A line portion of the vehicle body of the vehicle is detected based on the image, a height of the detected plurality of line portions is calculated, and a surface sandwiched between the two line portions having substantially the same height is defined as a horizontal plane of the vehicle body of the vehicle. Extracting, and
Determining the color of the extracted horizontal plane based on the image;
And a step of discriminating the color of the vehicle body based on the discriminated color of the horizontal plane. A line portion of the vehicle body of the vehicle is detected based on a photographed image obtained by photographing a vehicle traveling on a road, and the height and length of the detected plurality of line portions are calculated. The surface between the two line parts is extracted as the horizontal plane of the vehicle body of the vehicle,
Discriminate the color of the extracted horizontal plane based on the captured image,
A vehicle body color determination method, wherein the vehicle body color is determined based on the determined horizontal plane color. A program executed by a computer,
Capture an image of a vehicle traveling on the road,
A line portion of the vehicle body of the vehicle is detected based on the image, the heights of the detected plurality of line portions are calculated, and a surface sandwiched between two line portions having substantially the same height and substantially the same length is obtained. Extracting as a horizontal plane of the body of the vehicle;
Determining the color of the extracted horizontal plane based on the image;
And a step of discriminating the color of the vehicle body based on the discriminated color of the horizontal plane.

Images