JP2000222673A - Vehicle color discriminating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To discriminate a vehicle without using the image recognition of a number plate by discriminating the color of the vehicle based on color information in a discrimination area specified based on concentration distribution in an image. SOLUTION: A camera 4 always image-picks up the image of a prescribed position on an invasion road and outputs a video signal 7. In a vehicle color identification device 6, the video signals are A/D-converted by the camera interface part 611 of an image input part 61 and they are sequentially stored in an image memory 612. A vehicle detection part 5 detects that the vehicle arrives and outputs a vehicle detection signal 8 as a trigger for a processing. The vehicle position detection part 621 of a discrimination area stetting part 62 selects and reads the image stored in the image memory 612. A discrimination area setting part 623 detects the painting face of a hood and the like as the painting discrimination part of the vehicle and as an object and cuts a prescribed area as the discrimination area. The color identification part 631 of a vehicle color discrimination part 63 selects the average luminance value of the respective components of R, G and B in the discrimination area and a corresponding color from a color classification table 632 and outputs them.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle color discriminating apparatus for judging a paint color of a vehicle by using an image processing technique.

[0002]

2. Description of the Related Art FIG. 25 is a conceptual diagram of a vehicle identification device conventionally used for entrance / exit management of a parking lot or a management area.
In FIG. 25, 1 is an approach road, 2 is an approaching vehicle,
Reference numeral 3 denotes a support installed on the approach road and for attaching the monochrome camera 4b. Reference numeral 5 denotes a vehicle detection unit that detects arrival of the vehicle, and reference numeral 17 denotes a number recognition unit that determines a number written on a license plate of the vehicle from an image signal from the camera 4b.

FIG. 26 shows an example of the configuration of a conventional device, and an outline of the operation will be described below. In FIG. 26, first, a vehicle detection unit 5 is configured by a photoelectric switch or the like, detects that the vehicle has passed an area captured by the camera 4b, and outputs a vehicle detection signal 8, which is a pulse signal indicating vehicle detection, to the camera interface unit 611. Output to The camera interface unit 611 receives the pulse signal of the vehicle detection and outputs a photographing trigger signal 10 to the camera 4b, and the camera 4b uses the photographing trigger signal 10 as a trigger to photograph the vehicle and transmit a video signal 7 to the camera interface unit. 611
Output to Further, the video signal 7 received by the camera interface unit 611 is A / D converted, and
2 is stored. The number recognizing unit 17 reads out the image data stored in the image memory 612, recognizes the number from the distribution of black pixels and white pixels in the image data according to a predetermined procedure, and outputs the result to the recognition result output unit 69.
Output. As described above, the conventional vehicle identification device identifies the vehicle by recognizing the license plate number.

[0004]

In the conventional vehicle identification device, the vehicle is identified only by the license plate. Therefore, if some characters of the number cannot be recognized due to bending or dirt of the plate, the vehicle is reliably identified. There was a problem that it was not possible.

In order to solve such a problem, the vehicle color discriminating apparatus according to the present invention can identify a vehicle using a vehicle identification method other than the conventional vehicle identification method using license plate image recognition. The goal is to get what you can. It is another object of the present invention to improve the identification accuracy of vehicle identification using image recognition of a license plate.

[0006]

According to a first aspect of the present invention, there is provided a vehicle color discriminating apparatus which is installed on a road and acquires a color image, a vehicle detecting means for detecting a vehicle entering the visual field of the camera, Determination area setting means for extracting a determination area of an image corresponding to a specific part of the vehicle based on a density distribution in an image acquired by the camera in response to detection of the vehicle by the vehicle detection means; Color identification means for identifying the color of the vehicle based on the color information in the image in the determination area extracted by the area setting means.

[0007] A vehicle color discriminating apparatus according to a second aspect of the present invention includes:
In the first invention, the color identification means includes a color image of R (red), G (green), B
(Blue) The average luminance value of each component is calculated, and from each of the RGB average luminance values, a black-and-white determination unit for identifying any of white, black, and chromatic colors, and the black-and-white determination unit identifies the color as a chromatic color. In this case, there is provided a color determining means for determining a color name based on a chromaticity diagram corresponding to the average luminance value of RGB.

According to a third aspect of the present invention, in the vehicle color identifying apparatus according to the first aspect, the color identifying means includes an R (red), G (green), and B (blue) component of a color image in the identifying area. And a black-and-white judging means for discriminating any of white, black, and chromatic colors from each of the RGB average luminance values, and each of the RGB colors when the black-and-white judging means identifies the color as a chromatic color. Color determination means for determining a relative ratio of the average luminance value and determining a color name based on a root-mean-square value of a difference between each relative ratio and a preset reference value.

According to a fourth aspect of the present invention, in the vehicle color discriminating apparatus according to the first aspect, the color discriminating means includes an R (red), G (green), and B (blue) component of a color image in the discriminating area. Black-and-white determining means for determining the relative ratio of each average luminance value, calculating lightness from each relative ratio, and identifying one of white, black, and chromatic colors based on the calculated lightness; and Hue, saturation is calculated from the relative ratio of the respective average luminance values when identified as a chromatic color by the means, the hue of the reference color, the calculated hue based on a color classification table storing the saturation in advance, Color determination means for determining a color name corresponding to saturation.

According to a fifth aspect of the present invention, in the vehicle color determining apparatus according to the fourth aspect, the color determining means includes a gray determining section for distinguishing gray based on the calculated saturation.

According to a sixth aspect of the present invention, in the vehicle color discriminating apparatus according to the first aspect, the vehicle detecting means includes means for extracting a license plate from an image obtained by the camera and detecting entry of the vehicle. It is.

According to a seventh aspect of the present invention, in the vehicle color identification apparatus according to the first aspect, the identification area setting means extracts an identification area corresponding to a specific part of the vehicle as an area in which a plurality of small areas are combined. The color identification unit identifies a corresponding color name based on the color information for each of the plurality of small areas extracted by the determination area setting unit, and for each of the identified color names, Means for calculating the number of small areas having colors and comparing the calculation result with a predetermined value to determine the color of the vehicle which is painted in a plurality of colors.

According to an eighth aspect of the present invention, there is provided a vehicle color determining apparatus according to the first aspect, further comprising an illuminating section which emits visible light in synchronization with a shutter timing of the camera.

According to a ninth aspect of the present invention, there is provided the vehicle color discrimination device according to the first aspect, wherein the image information of the discrimination area extracted by the discrimination area setting means is transmitted together with the color information identified by the color discrimination means. And a display unit that receives the information transmitted from the image transmission unit and displays the received color information and image information.

According to a tenth aspect of the present invention, there is provided a vehicle color determining apparatus comprising:
In the invention according to the invention, an image transmitting unit for transmitting an RGB value in an image of the discrimination area extracted by the discrimination area setting unit together with the color information identified by the color identification unit, and receiving information transmitted from the image transmission unit The image processing apparatus further includes an image in which the original colors are reproduced from the received RBG values, and a display unit that displays color information.

According to an eleventh aspect of the present invention, there is provided a vehicle color determining apparatus comprising:
In the invention, the license plate of the vehicle output from the license plate recognition means for detecting a license plate of the vehicle from an image acquired by the camera and reading the number of the detected license plate is detected. A collation device that collates the number information and the vehicle color information obtained from the color identification unit with a pre-registered vehicle, and outputs a coincidence signal when they match.

According to a twelfth aspect of the present invention, there is provided a vehicle color determining apparatus comprising:
In the invention according to the invention, a license plate detecting means for detecting a license plate of the vehicle from an image acquired by the camera, and a distance measuring means for measuring a distance to the vehicle in accordance with the detection of the license plate detecting means, The size or shape of the vehicle is measured from the time change of the distance to the vehicle measured by the distance measuring means, and the type of the vehicle is determined by comparing the measurement result with a predetermined size or shape corresponding to the vehicle. Vehicle type discriminating means, license plate recognizing means for reading the number of the license plate detected by the license plate detecting means, vehicle type discriminated by the vehicle type discriminating means, and the number information of the vehicle output from the license plate recognizing means And the vehicle color information obtained from the color identification unit is compared with a vehicle registered in advance, Those having a matching device that outputs a coincidence signal to the multiplexer.

According to a thirteenth aspect of the present invention, there is provided a vehicle color determining apparatus comprising:
In the invention of the above, image transmitting means for sequentially transmitting the images captured by the camera, and receiving means for receiving the image transmitted by the image transmitting means, and outputting the received image to the determination area setting means It is provided.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 1 and 2
1 and 2 are an external view and a configuration diagram, respectively, showing Embodiment 1 of the present invention. In FIGS. 1 and 2, 1 to 3 have the same functions as those of the conventional example shown in FIG. Reference numeral 2 denotes an incoming vehicle, reference numeral 3 denotes a support installed on the approach road, reference numeral 4 denotes a camera mounted on an arm of the support 3 for capturing a color image of a passing vehicle, and reference numeral 5 denotes an imaging position of the camera 4. A vehicle detection unit 6 detects that the vehicle has arrived at the vehicle. A video signal from the camera 4 is input to the vehicle detection unit 6 to identify the color of the vehicle. The vehicle color identification device 6 further includes an image input unit 61 for inputting and storing an image output from the camera 4, and an area for determining a paint color stored in the image input unit 61 (hereinafter, referred to as a determination area). A determination area setting unit 62 to be set, a vehicle color determination unit 63 that determines the color of the determination area set by the determination area setting unit 62,
An output unit 64 that outputs the result determined by the vehicle color determination unit 63
Consists of

An example of the operation of the vehicle color identification device shown in the first embodiment will be described below. In the above device, the camera 4 always captures an image at a predetermined position on the approach road and outputs a video signal 7. In the vehicle color identification device 6, first, the video signal is A / D converted by the camera interface unit 611 of the image input unit 61, and is sequentially stored in the image memory 612.
The vehicle detection unit 5 detects that the vehicle has arrived at the imaging position of the camera 4, and outputs a vehicle detection signal 8 as a trigger for performing processing. The vehicle position detection unit 621 of the determination area setting unit 62 selects and reads the image stored in the image memory 612 from the image memory 612 as the processing target image when the vehicle detection signal 8 from the vehicle detection unit 5 is received. In the present embodiment, the image is continuously input, and the vehicle is specified by ANDing with the vehicle detection signal 8 later. However, the camera 4 reaches the photographing point with the vehicle detection signal 8 as a trigger. It is also possible to capture only one image of the vehicle.

Next, an example of an image processing method for setting a determination area will be described. First, the vehicle position detecting section 621 extracts the reference image memory 62 in order to extract only the vehicle from the image.
2, a reference image which is a background image when there is no vehicle, and an image memory 612 as a processing target image.
The general position of the vehicle as a moving object is specified by using a general image processing technique such as an inter-image difference for an image selected and read out from (a) and (b) in FIG. The discrimination area setting unit 623 obtains a density distribution of one or more lines in the vertical direction of the vehicle position of the input image in order to detect a painted surface such as a bonnet as a part for discriminating the paint color of the vehicle ( FIG. 3 (c)). Since the density distribution of a bonnet or the like is generally uniform, its position is specified as a portion having a low density change rate. Next, a density distribution is obtained in the horizontal direction with respect to the position of the hood and the like detected in the vertical direction (FIG. 3D), and the horizontal direction of the position of the bonnet and the like is specified by the same method. Further, a determination area setting unit 623
In the above, a certain area is cut out of the determined area for actually determining the paint color within the position of the specified hood or the like. For example, in the case where the imaging range of the camera is about 2 m, the vehicle occupies a large area on the screen, and the vehicle shooting position on the screen is stabilized by the vehicle detection signal 8 from the vehicle detection unit 5. , Because most of the screen is occupied by the vehicle and is hardly affected by the background including the road during processing.
The process of detecting the vehicle position in the vehicle position detector 621 using the reference image memory 622 can be omitted.

In the color discriminating section 631 of the car color discriminating section 63, first, the average luminance of each component of R (red), G (green), and B (blue) in the discriminating area set by the discriminating area setting section 62. Find the value. In the color classification table 632, R (red), G
A correspondence table between each of the luminance values of (green) and B (blue) and the color name is set, and the color identification unit 631 refers to the color classification table 632 for the name of the color corresponding to the calculated average luminance value. And outputs the result via the output unit 64.

As described above, according to this embodiment, one of the features of the vehicle completely different from the license plate number is as follows.
One of the colors of the vehicle can be identified. Conventionally, when identifying a vehicle by license plate number recognition, if some characters cannot be recognized, it will be specified only from readable characters, so the target range will be expanded and it will not be possible to completely identify it. In such a case, the target range can be narrowed by including the information of the vehicle color in the judgment item, and the specific accuracy can be improved. Furthermore, when performing collation with a vehicle that has been registered in advance, such as a parking lot or a facility management facility, by comparing both the number and color information with the registered information,
Even if the numbers match, it is possible to take a method of not recognizing the registered vehicle if the colors do not match, and to check if the number is incorrectly recognized, thereby improving the reliability.

Embodiment 2 FIG. In the invention according to the second embodiment, the color identification unit 631 shown in FIG. 2 has a determination circuit for performing determination using an algorithm as shown in FIG. FIG. 4 shows only the processing content of the color identification unit 631, and the configuration and operation of the other parts are the same as those in FIG. First, the average luminance of each of the R (red), G (green), and B (blue) components in the paint color determination area is determined by the black and white determination means of the color identification unit 631 (S41). Next, the sum T of the R (red), G (green), and B (blue) components is calculated (S42). Since T indicates the brightness of the whole image, first, when T is larger than the white level determination value W, the image is determined to be white (S43), and when T is smaller than the black level determination value B, it is determined to be black (S44). Next, the color identification unit 631
R (red), G (green), B
(Blue) is subjected to xy conversion (S45). xy conversion is CIE
(International Commission on Lighting, Commission Intermission
X = 0.49000R + 0.31000G + O.D. using the method defined in “National d'Eclairage”. 200
00B Y = 0.17697R + 0.81240G + 0.010
63B. Further, in the color determination means, the color classification table 63
As for 2, a classification table having the same format as the chromaticity diagram defined by CIE as schematically shown in FIG. 5 is used, and the xy values calculated above are applied to FIG. Is selected (S46). However, even if the same color is used for photographing an outdoor vehicle, there is a difference in the appearance depending on the environmental conditions, and it is difficult to make a fine distinction. Therefore, as shown in FIG. Classify.

Embodiment 3 In the invention according to the third embodiment, a circuit is provided in which the color identification unit 631 shown in FIG. 2 makes a determination using an algorithm as shown in FIG. In the black-and-white determination means of the color identification unit 631, white and black are determined by the method shown in FIG. 4 (S41 to S44) of the second embodiment (corresponding to S61 to S64 in FIG. 6). If not, the relative ratios r, g, and b of the R (red), G (green), and B (blue) components of the image in the area set by the determination area setting unit 62 are obtained (FIG. 6). S65). On the other hand, in the color classification table 632, r _i , g _i , and b _i values corresponding to a plurality of reference colors are stored in advance as shown in FIG. The color determination means calculates the root mean square RMS _i of the difference between the previously obtained relative ratio r, g, b of the color component of the input image and the reference value based on Equation 1 (S66 in FIG. 6), and calculates the value. Is determined as the paint color of the target vehicle (S67 in FIG. 6).

[0026]

(Equation 1)

Embodiment 4 The invention according to the fourth embodiment has a circuit that makes a determination in the color identification unit 631 shown in FIG. 2 using an algorithm as shown in FIG. First, the average brightness of each of the R (red), G (green), and B (blue) components in the paint color determination area is obtained by the black and white determination means of the color identification unit 631 (S81). Next, R (red), G (green), B (blue)
The relative ratios r, g, and b of the respective components are determined (S82). Further, the hue (H), saturation (S), and lightness (V) are calculated from the relative ratios r, g, and b of the input image using HSV conversion (S
83) First, white and black are determined based on the lightness (V) (S8).
4, S85), and then in the color determination unit of the color identification unit 631,
The gray is determined by the saturation (S) (S86), and finally the color name corresponding to the hue H is determined by the conversion table based on the Marcel hue circle as shown in FIG. S87). Some equations are already known for the HSV conversion. For example, the following equation according to the definition of Raines is used.

[0028]

(Equation 2)

The hue conversion table is a color classification table 632.
Is stored.

Embodiment 5 In the invention according to the fifth embodiment, the color identification section 631 shown in FIG. 2 includes a gray determination section that performs determination using an algorithm as shown in FIG. First, the average brightness of each of the R (red), G (green), and B (blue) components in the paint color determination area is obtained by the black and white determination means of the color identification unit 631 (S101). Next, R (red), G (green), B
The relative ratios r, g, and b of each component of (blue) are determined (S10
2). Further, the hue (H), saturation (S), and brightness (V) are calculated from the relative ratios r, g, and b of the input image using HSV conversion (S103), and white and black are determined based on the brightness (V). (S104, S105). Next, the color determination unit of the color identification unit 631 calculates the hue (H), and determines the color name by using a conversion table having the same format as that shown in FIG. In general, even if the actual subject is gray due to the characteristics of the camera 4, the camera output may be biased to any color. Therefore, only when a specific color name is determined based on the hue determination, the saturation (S ) To determine whether the color is gray. For example, RGB
Of the cameras with strong B (blue), even if the color is gray, the RGB balance is lost and the color is determined to be blue. Therefore, once the color is determined to be blue (S107), the distinction between blue and gray is made based on the saturation (S). (S108).

Embodiment 6 FIG. FIG. 11 shows a configuration diagram of a vehicle identification device of the invention according to the sixth embodiment. In FIG. 11, a license plate detection unit 624 sequentially reads out images continuously input from the camera 4 via the camera interface unit 611 from the image memory 612, and detects a license plate. The method of detecting a license plate uses a conventionally used method. For example, a binary image is obtained by binarizing a luminance signal of an image read from the image memory 612. The presence / absence of a plate is determined by matching the distribution of white and black pixels in the binary image with the distribution of white and black pixels in the character array of a known license plate. If it is determined in this process that there is a plate, the image is sent to the determination area setting unit 623 as an image to be processed. The area around the number is a portion where the brightness of the image is greatly changed, such as a front grill or a bumper of the vehicle. However, there is usually a region above the number between the windshield and the passenger regardless of the passenger car or truck, where the paint color is stable. As shown in FIG. 12, the discrimination area setting unit 623 searches for a change in luminance upward from the license plate because the license plate position is known. The method of setting the paint color determination area is the same as the method described in the first embodiment. According to this operation, since the plate position is known, the paint color determination area can be set relatively easily. After the paint color discriminating area is set, the color discriminating unit 631 first sets R (red) and G in the area set by the discriminating area setting unit 62, as in the first embodiment.
The average luminance value of each component (green) and B (blue) is determined. In the color classification table 632, R (red), G (green), B
A correspondence table between each luminance value of (blue) and a color name is set, and the color identification unit 631 selects a color name by referring to the color classification table 632 based on the obtained average luminance value, and outputs it to the output unit 64. Output the result via. According to the vehicle color discriminating apparatus of this embodiment, the vehicle detection unit is not required, so that the cost can be reduced and the camera can be installed only. Further, since it is known that the target screen includes the vehicle, it is possible to avoid an erroneous detection of determining an object other than the vehicle.

Embodiment 7 FIG. 13 shows the configuration of the invention according to the seventh embodiment, and FIG.
FIG. 15 shows a flow of processing from setting of a plurality of areas in the determination area setting unit 625 for multicolor determination to color name determination in the color identification unit 631. 13 and 14, first, the vehicle position detection unit 621 specifies the position of the vehicle by the method described in the first embodiment. Next, the determination area setting unit 625 for multicolor determination sets a predetermined number N of small determination areas in the area determined to be a bonnet as shown in FIG. 14E (S151 in FIG. 15). However, when specifying the hood, if the vehicle is painted with two or more paint colors, the entire hood will not have a uniform brightness distribution,
A portion where there is no luminance variation over a predetermined length is defined as a bonnet. Next, the color identification unit 631 refers to the color classification table 632 for all of the discrimination areas by the same method as that described in the first embodiment.
Color determination is performed individually as shown in FIG. 4 (f) (S15 in FIG. 15).
2 to S153). As a result of the color determination of all areas, if the ratio ni / N of the number ni of the determination areas determined to be a certain color number i to the whole is smaller than the set value, it is excluded as a part not related to the paint color such as decoration. Ni as in 14 (g)
Only those for which / N is greater than or equal to the set value are valid (S1 in FIG. 15).
54-S158). As a result of this processing, only the color name corresponding to the valid color number is output from the output unit 64 (FIG. 1).
S159 of No. 5). This makes it possible to identify the color of the vehicle painted in a plurality of colors, such as two-tone colors, and to catch the entire color even when a pattern or character is displayed in a part.

Embodiment 8 FIG. FIG. 16 shows an external view of a vehicle color discrimination device of the invention according to the eighth embodiment. In FIG. 16, 1 to 6 are the same as those shown in FIG. Reference numeral 9 denotes an illumination device synchronized with the shutter of the camera 4. The purpose of the present vehicle color determination device is to determine the paint color of the vehicle, so that a light source such as xenon, which is close to the spectrum of sunlight, is used for illumination. Reference numeral 25 denotes a photoelectric sensor that converts ambient brightness into an electric signal, and outputs a light amount signal 26. FIG. 17 shows the signal flow. 17, 3 to 8 are the same as those shown in FIG. Camera interface section 611
Then, upon receiving the vehicle detection signal 8 from the vehicle detection unit 5, it outputs an external synchronization shutter command 10 to the camera 4. At this time, when the light amount signal 26 is equal to or less than a preset value, it is determined that the light amount is insufficient, and the light emission command 11 is output to the illumination device 9 simultaneously with the shutter command 10, and the camera 4
, And is stored in the image memory 612. The video signal 7 is an image at the time of light emission, and an image having sufficient luminance can be obtained even in a place where originally there is little visible light such as an underground parking lot or an outdoor night. The determination area setting unit 623, the color identification unit 631, and the color classification table 632 have the same functions as those described in the first embodiment, perform the same operations as in the first embodiment, determine colors, and output. 64. Further, in this embodiment, visible light is emitted from the lighting 9, but at this time, the lighting 9 is installed at a position where the driver is not affected by dazzling or the like. That is, when the target vehicle comes to the shooting position,
In order to prevent the emitted light from reaching the driver's eyes directly, the installation position of the lighting 9 is set at a position where the sensor detects the vehicle, near a position immediately behind the sensor, or on the roadside. When irradiating from behind, the camera 4 is set at a position where regular reflection does not occur so as not to cause halation on the camera 4. Furthermore, lighting 9
Attach a diffuser to the front of the device to prevent uneven illuminance. This makes it possible to identify the color of the vehicle even at night or in a dark area indoors.

Embodiment 9 FIG. 18 shows a configuration diagram of a vehicle color determination device according to the ninth embodiment. In FIG. 18, steps 3 to 8 are the same as those in FIG. Reference numeral 12 denotes an image compression unit for compressing image data of an area used for discrimination, and reference numeral 13 denotes a transmission memory for temporarily storing an image compressed by the image compression unit 12 and an acquisition time of the image. It consists of memory. Further, an image transmission unit 65 transmits the result of the color determination associated with the image based on the acquisition time and the image at that time via a private line or the like. Also one
Reference numeral 4 denotes a display device which is installed at another point and displays an image transmitted from the image transmission unit 65 and a color name as a result of the automatic determination on a display or the like, and allows an operator to compare the color name and the image through the display. . The vehicle color discriminating apparatus according to the ninth embodiment is configured as described above. The image input from the camera 4 to the image memory is immediately compressed by the image compressing unit 12, the capacity is temporarily reduced and the image number is temporarily stored. By managing, the result can be transmitted together with the input image after completion of the automatic determination of the paint color with a small memory capacity. As a result, the operator can visually confirm whether the result of the automatic determination is correct or not at a remote location, and can also correct the result of the automatic determination.

Embodiment 10 FIG. FIG. 19 shows a configuration diagram of a vehicle color determination device according to the tenth embodiment of the invention. In FIG. 19, reference numeral 15 denotes R in the discrimination area used by the color identification unit 631.
RGB that temporarily stores (red), G (green), and B (blue) information
The storage unit 66 is an RGB transmission unit that transmits both the RGB values of the determination area temporarily stored in the RGB storage unit 15 and the color names automatically determined by the color identification unit 631 using the values.
16 is installed at another point, reproduces and displays the original color from the RGB information transmitted from the RGB transmission unit 66,
This is an image reproducing apparatus that can compare the results of the automatic determination. The vehicle color discrimination device according to the tenth embodiment is configured as described above, and compresses, saves, and transmits the image information used for the paint color discrimination into RGB values, so that a smaller amount of memory is used as compared with the ninth embodiment. Depending on the line capacity, it is possible to construct a system in which the operator can visually confirm the correctness of the result of the automatic determination at a remote place and correct the result of the automatic determination.

Embodiment 11 FIG. FIG. 20 is an external view of a vehicle identification device of the invention according to Embodiment 11, and FIG. In FIG. 21, the license plate detecting unit 624 sequentially reads out images continuously input from the camera 4 via the camera interface unit 611 from the image memory 612 as in the sixth embodiment, and The license plate is detected by using the pattern matching technique used in the vehicle number reading device of No. 1. If it is determined in this process that there is a plate, the image is sent to the determination area setting unit 623 as an image to be processed. On the other hand, the image of the detected license plate portion is transmitted to the license plate recognition unit 17, and is used in a conventional vehicle number reading device such as a pattern matching method for matching each character of the plate image with a reference character pattern registered in advance. The number displayed on the license plate is recognized using the technique used. The identification result of the vehicle color and the identification result of the number are output together from the paint color and number transmission unit 67. The identification result of the vehicle color and the recognition result of the number output from the number transmission unit 67 are collated with the vehicle color / number of the vehicle registered in advance by a collation device (not shown). Is done. In the vehicle color discriminating apparatus of this embodiment, since the number information and the paint color information of the vehicle are known, the accuracy of specifying the vehicle is improved.

Embodiment 12 FIG. FIG. 22 is an external view of the vehicle color discriminating apparatus according to the twelfth embodiment, and FIG.
Shown in 22 and 23, 1 to 63 are the same as those in FIG. 2. In FIG. 22, as in the sixth embodiment, the license plate
, Sequentially read from the image memory 612 via the camera interface unit 611, and detects a license plate using a pattern matching technique used in a conventional vehicle number reading device.
If it is determined in this process that there is a plate, the image is sent to the determination area setting unit 623 as an image to be processed. On the other hand, the image of the detected license plate portion is transmitted to the license plate recognition unit 17, and is used in a conventional vehicle number reading device such as a pattern matching method for matching each character of the plate image with a reference character pattern registered in advance. Recognized using the technique used. Further, reference numeral 18 denotes a light projector provided at the upper part of the approach road 1 and which emits light in accordance with the detection of the license plate by the license plate detection unit 624. Reference numeral 19 denotes the height of the vehicle body in accordance with the traveling of the vehicle from the position of the reflected light. A distance measuring unit that measures the time change of the vehicle and measures the cross-sectional shape and dimensions of the vehicle in the traveling direction of the approach road 1; 20 is a vehicle type determining device that determines the type of a vehicle sedan, one box, or the like from the cross-sectional shape of the vehicle A transmission line 68 transmits the result of the vehicle color, the number, and the type of the vehicle together. In the vehicle color discriminating apparatus according to this embodiment, since the vehicle type discriminating apparatus 20 can know the size and cross-sectional shape of the vehicle in addition to the number information and the paint color information shown in the eleventh embodiment, The accuracy of the identification is improved. Note that the distance measurement by the distance measurement unit 19 may be horizontal, for example, as long as the distance to the vehicle in a direction perpendicular to the traveling direction of the vehicle is measured. In this case, the vehicle length is measured. it can.

Embodiment 13 FIG. FIG. 24 shows a configuration diagram of a vehicle color discrimination device according to the thirteenth embodiment of the invention. In FIG. 24, reference numeral 4 denotes a camera for capturing an image of a vehicle, reference numeral 5 denotes a vehicle detection unit for detecting arrival of the vehicle, reference numeral 21 denotes a data compression unit for compressing only an image at the time of vehicle detection, and reference numeral 22 denotes a transmission unit for transmitting a compressed image. is there. The central unit 23 is installed at a remote location and receives and processes data from a plurality of transmission units 22. An image input unit 24 receives image data from the transmission unit 22. Further, the discriminating area setting unit 62 and the car color judging unit 63 for performing the car color discrimination have the same functions as those shown in FIG. 2, and the discriminating area setting unit 62 includes a vehicle in the image. Therefore, a determination area for determining the paint color of the vehicle from the corresponding image is set in the same manner as in the first embodiment. In the color identification section 631, the color classification table 6 set in advance based on the RGB ratio and size of the image in the area set by the determination area setting section is set.
32, the color name is determined, and the result is output via the output unit 64. According to the vehicle color identification apparatus of the thirteenth embodiment, it is not necessary to have a vehicle color identification function for each camera. In addition, by having a relatively large capacity memory in the central unit, it is possible to perform paint color discrimination off-line or to perform paint color discrimination only for the designated camera, enabling highly flexible operation. Becomes

The inventions according to the first to thirteenth embodiments can be applied to other systems other than a parking lot or a specific facility, as long as the system identifies a vehicle.

[0040]

According to the first aspect of the present invention, it is possible to obtain information on a vehicle paint color based on a color image. by this,
It is possible to obtain identification information of a vehicle different from character information of a license plate conventionally used.

According to the second aspect, it is possible to obtain color information along an existing chromaticity diagram.

According to the third aspect, since the color is specified by the difference from the reference color, it is possible to obtain a judgment result corresponding to a preset color classification without complicated conversion.

According to the fourth aspect, color information along the Munsell hue circle can be obtained.

According to the fifth aspect of the present invention, by obtaining the saturation, when the gray color approaches a specific color due to a color shift of the camera, the gray color can be identified.

According to the sixth aspect, since the vehicle is directly detected from the image and the paint color discrimination area is set, a vehicle detecting means other than the camera is not required, which leads to cost reduction and improvement in installation flexibility.

According to the seventh aspect, by setting a plurality of small areas in the discrimination area, it is possible to identify a vehicle painted in a plurality of colors.

According to the eighth aspect, since illumination synchronized with the shutter is provided, it can be used at night or in a dark place underground.

According to the ninth aspect, since the target image is transmitted together with the determined color name information, it is possible to confirm and correct the correctness of the color name at another point.

According to the tenth aspect, since the RGB information is transmitted together with the determined color name information, it is possible to confirm and correct the correctness of the color name by combining the actual colors at other points.

According to the eleventh aspect, since the vehicle color identification device and the license plate recognizing means are combined, even if some characters cannot be recognized by the number reading, the paint color information is referred to, and the specific accuracy of the vehicle can be improved. improves.

According to the twelfth aspect, in order to have the vehicle color identification device, the number reading device and the vehicle type identification device together,
Even when some characters cannot be recognized by reading a predetermined number (seven numbers), the vehicle identification accuracy can be improved by referring to the paint color information and the vehicle type (vehicle shape).

According to the thirteenth aspect, since the vehicle color can be identified at another point by transmitting an image, it is not necessary to provide a vehicle color identification function on the road side, so that the cost can be reduced. Processing can be performed for a specific vehicle as needed, such as when the information is inadequate.

[Brief description of the drawings]

FIG. 1 is a first embodiment of a vehicle color determination device according to the present invention;
It is a key map showing appearance of.

FIG. 2 is a first embodiment of a vehicle color discriminating apparatus according to the present invention;
FIG. 3 is a diagram showing the configuration of FIG.

FIG. 3 is a first embodiment of a vehicle color determination device according to the present invention;
It is a figure which shows operation | movement.

FIG. 4 is a second embodiment of the vehicle color discriminating apparatus according to the present invention;
It is a figure explaining operation of.

FIG. 5 is a second embodiment of the vehicle color discriminating apparatus according to the present invention;
FIG. 3 is a diagram illustrating an outline of a chromaticity diagram shown in FIG.

FIG. 6 is a third embodiment of the vehicle color discriminating apparatus according to the present invention;
It is a figure explaining operation of.

FIG. 7 is a third embodiment of the vehicle color discriminating apparatus according to the present invention;
FIG. 4 is a diagram for explaining an example of a color classification table of FIG.

FIG. 8 is a fourth embodiment of the vehicle color discriminating apparatus according to the present invention;
It is a figure explaining operation of.

FIG. 9 is a fourth embodiment of the vehicle color discriminating apparatus according to the present invention;
FIG. 4 is a diagram for explaining an example of a color classification table of FIG.

FIG. 10 is a diagram illustrating an operation of a vehicle color determination device according to a fifth embodiment of the present invention.

FIG. 11 is a diagram showing a configuration of a vehicle color determination device according to a sixth embodiment of the present invention.

FIG. 12 is a diagram illustrating the operation of a vehicle color discriminating apparatus according to a sixth embodiment of the present invention.

FIG. 13 is a diagram showing a configuration of a vehicle color determination device according to a seventh embodiment of the present invention.

FIG. 14 is a diagram illustrating the operation of a vehicle color discriminating apparatus according to a seventh embodiment of the present invention.

FIG. 15 is a diagram illustrating the operation of the vehicle color discriminating apparatus according to the seventh embodiment of the present invention.

FIG. 16 is a conceptual diagram showing an appearance of a vehicle color identification device according to an eighth embodiment of the present invention.

FIG. 17 is a diagram showing a configuration of a vehicle color identification device according to an eighth embodiment of the present invention.

FIG. 18 is a diagram showing a configuration of a ninth embodiment of a vehicle color determination device according to the present invention.

FIG. 19 is a diagram showing a configuration of a tenth embodiment of a vehicle color determination device according to the present invention.

FIG. 20 is a conceptual diagram showing an appearance of an eleventh embodiment of a vehicle color determination device according to the present invention.

FIG. 21 is a diagram showing a configuration of a vehicle color identification device according to an eleventh embodiment of the present invention.

FIG. 22 is a conceptual diagram showing an appearance of a twelfth embodiment of the vehicle color determination apparatus according to the present invention.

FIG. 23 is a diagram showing a configuration of a twelfth embodiment of a vehicle color determination device according to the present invention.

FIG. 24 is a conceptual diagram showing a configuration of a thirteenth embodiment of a vehicle color determination device according to the present invention.

FIG. 25 is a conceptual diagram showing the appearance of a conventional vehicle color determination device.

FIG. 26 is a diagram showing a configuration of a conventional vehicle color determination device.

[Explanation of symbols]

1 Approach road, 2 target vehicles, 3 pillars, 4 cameras, 4
b B / W camera, 5 vehicle detector, 6 vehicle color discriminator, 7
Video signal, 8 vehicle detection signal, 9 lighting, 10 shooting trigger signal, 11 light emission command, 12 image compression unit, 1
3 transmission memory, 14 display device, 15 RGB storage unit, 16 image reproduction device, 17 license plate recognition unit, 18 floodlight, 19 distance measurement unit, 20 vehicle type discrimination device, 21 data compression unit, 22 transmission unit, 23 central unit, 24 image input unit, 25 photoelectric sensor, 26 light intensity signal, 61 image input unit, 62 discrimination area setting unit, 6
3 car color judgment unit, 64 output unit, 65 image transmission unit, 66
RGB transmission section, 67 paint number transmission section, 68 paint number vehicle type transmission section, 68 recognition result output section, 611
Camera interface unit, 612 Image memory, 6
21 vehicle position detecting section, 622 reference image memory, 62
3 Discrimination area setting unit, 624 license plate detection unit, 625 multicolor judgment discrimination area setting unit, 631 color identification unit, 632 color classification table.

Continued on front page F-term (reference) 5B057 CH14 DA07 DA08 DA12 DB06 DC02 DC09 DC22 DC25 DC30 5H180 AA01 CC05 CC24 DD02 EE07 5L096 AA02 BA04 CA03 EA35 FA15 FA32 FA64 FA66 GA19 GA40 GA41 HA09 LA12 9A001 BB03 BB04 EE04H05 JJ77 KK16 KK37 KK42 KK60 LL02

Claims (13)

[Claims] 1. A camera installed on a road to acquire a color image, a vehicle detecting means for detecting the entry of a vehicle into the field of view of the camera, and the camera in response to detection of the vehicle by the vehicle detecting means Determination area setting means for extracting a determination area of an image corresponding to a specific part of the vehicle based on the density distribution in the image obtained in the step; and color information in the image in the determination area extracted by the determination area setting means And a color identification means for identifying the color of the vehicle based on the vehicle color. 2. The color identification means calculates an average luminance value of each of R (red), G (green), and B (blue) components of a color image in the discrimination area, and calculates each of the average luminance values of RGB. A black-and-white determining means for identifying any one of white, black, and chromatic colors, and a color name based on a chromaticity diagram corresponding to the average RGB value when the black-and-white determining means identifies a chromatic color 2. The vehicle color discriminating apparatus according to claim 1, further comprising: 3. The color identification means calculates an average luminance value of each of R (red), G (green), and B (blue) components of the color image in the discrimination area, and calculates each average luminance value of RGB. A black-and-white determining means for identifying any one of white, black, and chromatic colors, and a relative ratio of each average luminance value of RGB when the black-and-white determining means identifies the color as a chromatic color. 2. The vehicle color discriminating apparatus according to claim 1, further comprising: a color judging unit that judges a color name based on a root mean square of a difference from a set reference value. 4. The color discriminating means calculates a relative ratio of each average luminance value of each of R (red), G (green), and B (blue) components of the color image in the discrimination area. From the calculated brightness, white, black, based on the calculated brightness black and white determination means for identifying any of the chromatic color, and the black and white determination means when the chromatic color is identified by the average luminance value Hue and saturation are calculated from the relative ratios of the reference colors, and a hue of the reference color and color determination means for determining a color name corresponding to the saturation based on the color classification table storing the saturation in advance. The vehicle color discriminating apparatus according to claim 1, further comprising: 5. The vehicle color discriminating apparatus according to claim 4, wherein said color discriminating means includes a gray discriminating unit for discriminating gray based on the calculated saturation. 6. The vehicle color discriminating apparatus according to claim 1, wherein the vehicle detecting means includes means for extracting a license plate from an image acquired by the camera and detecting entry of the vehicle. 7. The discriminating area setting means includes means for extracting a discriminating area corresponding to a specific part of the vehicle as an area in which a plurality of small areas are combined, and the color discriminating means extracts the discriminating area by the discriminating area setting means. The corresponding paint color name is identified based on the color information for each of the plurality of small areas, and the number of the small areas having the color in the determination area is calculated for each of the identified color names, and the calculation result is obtained. 2. A vehicle color discriminating apparatus according to claim 1, further comprising means for comparing the color of the vehicle with a predetermined value to determine a color of the vehicle which is painted in a plurality of colors. 8. The vehicle color discriminating apparatus according to claim 1, further comprising an illumination unit that emits visible light in synchronization with a shutter timing of the camera. 9. An image transmission unit for transmitting image information of the discrimination area extracted by the discrimination area setting unit together with the color information identified by the color identification unit, and receiving the information transmitted from the image transmission unit. The vehicle color discriminating apparatus according to claim 1, further comprising a display unit for displaying the received color information and image information. 10. An image transmission unit for transmitting an RGB value in an image of a discrimination area extracted by the discrimination area setting unit together with the color information identified by the color identification unit, and information transmitted from the image transmission unit. 2. The vehicle color discriminating apparatus according to claim 1, further comprising a display unit that receives the image, reproduces an original color from the received RBG value, and displays color information. 11. A license plate recognizing means for detecting a license plate of the vehicle from an image acquired by the camera and reading the number of the detected license plate, and the vehicle output from the license plate recognizing means. 2. A vehicle color according to claim 1, further comprising: a matching device that checks the number information of the vehicle and the vehicle color information obtained from the color identification unit with a vehicle registered in advance, and outputs a match signal when the vehicle matches. Discriminator. 12. A license plate detecting means for detecting a license plate of the vehicle from an image acquired by the camera, and a distance measuring means for measuring a distance to the vehicle in accordance with the detection of the license plate detecting means. Measuring the size or shape of the vehicle from the time change of the distance to the vehicle measured by the distance measuring means, and discriminating the vehicle type by comparing the measurement result with a predetermined size or shape corresponding to the vehicle. Type discriminating means, a license plate recognizing means for reading the number of the license plate detected by the license plate detecting means, a vehicle type discriminated by the vehicle type discriminating means, and a number of the vehicle output from the license plate recognizing means. The information and the vehicle color information obtained from the color identification unit are compared with a pre-registered vehicle, and if there is a match, The vehicle color discrimination device according to claim 1, further comprising a matching device that outputs a coincidence signal when the vehicle color is determined. 13. An image transmitting means for sequentially transmitting images photographed by the camera, a receiving means for receiving an image transmitted by the image transmitting means, and outputting the received image to the determination area setting means. The vehicle color discriminating apparatus according to claim 1, further comprising: