JP2000193419A - Tire chain detecting device and studless tire detecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a tire chain detecting device and a studless tire detecting device wherein detecting of a tire chain attached car or studless tire attached car, which was visually inspected so far, is automatically performed by image processing. SOLUTION: The tire of a car is in an input image Ig through an image input part 11 and a image storage part 12. A square region image R13 for a tire part is extracted from the input image Ig by a judgement region extracting part 13. A tire pattern judging part 14 detects a shade part in the region image R13, and based on the shade, judges whether a tire chain is attached to the tire or not, a judgement result J14 is provided to an information presentation part 15. Based on the judgement result J14, the information presentation part 15 raises/lowers a crossing gate, for example.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for detecting the condition of a tire, and more particularly to a vehicle equipped with a tire chain and a studless tire (winter tire), which are non-slip tires for preventing slippage due to snow and the like. The present invention relates to a tire chain detecting device and a studless tire detecting device for detecting.

[0002]

2. Description of the Related Art Conventionally, the condition of tires has been inspected for the purpose of protecting the road surface in a tunnel or restricting traffic such as snow. These tires include chain mounted tires and studless tires. In long tunnels such as cold regions, to protect the road surface in the tunnel, check the mounting condition of the tire chain, and if you are wearing a tire chain,
Measures such as removing the chain are taken. In addition, at interchanges such as expressways and service areas, in order to regulate traffic due to snow, inspect the mounting condition of tire chains and studless tires.If these tire chains and studless tires are not installed, Measures will be taken, such as wearing chains and studless tires. However, these inspections are performed by visual inspection of a person, and there is no device for inspecting the condition of the tire on behalf of the person.

[0003]

The conventional inspection of the mounted state of a tire chain and the inspection of the mounted state of a studless tire are performed visually by a person going outdoors, so that at night or when the weather deteriorates, etc. There was a problem that working conditions deteriorated significantly. In addition, in an environment where the working conditions are deteriorated, there is a problem that the inspection accuracy is reduced and the inspection speed is also reduced. An object of the present invention is to provide a tire chain detection device and a studless tire detection device that automatically perform inspections conventionally performed by humans, and to alleviate working conditions or realize unmanned operation and improve inspection accuracy. And

[0004]

According to a first aspect of the present invention, there is provided a tire chain detecting apparatus for detecting a tire chain mounting state of a vehicle and presenting information of the detection result. In the following image input unit,
It comprises an image storage unit, a tire position detection unit, a determination region extraction unit, a chain attachment determination unit, and an information presentation unit. The image input unit includes an image input unit that captures an image of a predetermined place as an input image, and illumination. The image storage unit stores the input image. The tire position detector detects that the tire of the vehicle has arrived at the predetermined location, generates a detection signal, and indicates that the input image shows the tire. The determination region extraction unit is a unit that receives the detection signal and extracts a region image of a predetermined size including the tire from the input image stored in the image storage unit. The chain mounting determination unit is means for analyzing the extracted area image to extract a shadow formed by the illumination, and determining a tire chain mounting state of the vehicle based on the shadow. The information presentation section
The information presentation is performed based on the determination result of the chain attachment determination unit.

According to a second aspect of the present invention, there is provided a tire chain detecting device for detecting a mounted state of a tire chain of a vehicle and presenting information of the detection result, the following image input unit, image storage unit, tire position detecting unit, It is composed of a judgment area extraction unit, a chain attachment judgment unit, and an information presentation unit. The image input unit changes an imaging direction based on the given position information, and inputs an image taken by changing the zoom and focus based on the given position information as an input image. And illumination for changing the light irradiation direction based on the given position information. The image storage unit stores the input image. The tire position detector is means for detecting a tire position of the vehicle, sending the position information indicating the tire position to the image input unit, and generating a detection signal indicating that the tire position has been detected. The determination region extraction unit is a unit that receives the detection signal and extracts a region image of a predetermined size including the tire from the input image stored in the image storage unit.
The chain mounting determination unit is means for analyzing the extracted area image to extract a shadow formed by the illumination, and determining a tire chain mounting state of the vehicle based on the shadow. The information presenting section is for presenting the information based on the determination result of the chain mounting determining section.

According to a third aspect of the present invention, in the tire chain detecting device according to the first or second aspect, the determination region extracting section includes:
A contact surface of the tire is detected from a luminance hiss drum of the input image, and the rectangular region image of the tire portion in the input image is extracted based on the contact surface. According to a fourth aspect, in the tire chain detection device according to the third aspect, the determination area extracting unit moves a rectangle having the same size as the area image along the ground plane, and the luminance in the rectangle is minimized or reduced. Detect the position where the variance is minimized,
The rectangle at this position is extracted as the region image.

According to a fifth aspect of the present invention, in the tire chain detecting device according to the first or second aspect, the determination area extracting section extracts a rectangle at a predetermined position of the input image as the area image. ing. According to the first to fifth inventions, as described above, the tire chain detection device is configured, so that the input image in which the tire is captured is obtained by the image input unit, and the input image is obtained from the input image by the determination region extraction unit. An area image is extracted. The region image is analyzed by the chain mounting determination unit, and the mounting state of the tire chain of the vehicle is determined based on the shadow created by the illumination. Then, the information presentation unit presents information based on the determination result of the chain attachment determination unit.

According to a sixth aspect of the present invention, there is provided a studless tire detecting device for detecting a studless tire mounting state of a vehicle and presenting information of the detection result, the following image input unit, image storage unit, tire position detecting unit, Judgment area extraction unit,
The vehicle includes a tire pattern determination unit and an information presentation unit. The image input unit includes an image input unit that captures an image of a predetermined place as an input image, and illumination. The image storage unit stores the input image. The tire position detector is a unit that detects that the tire of the vehicle has arrived at the predetermined location, generates a detection signal, and indicates that the tire is captured in the input image. The determination region extraction unit is a unit that receives the detection signal and extracts a region image of a predetermined size including the tire from the input image stored in the image storage unit. The tire pattern determination unit analyzes the extracted area image, extracts fine grooves on the surface of the tire as characteristics of the studless tire, and determines a studless tire mounting state of the vehicle based on the characteristic. is there. The information presentation unit is for performing the information presentation based on the determination result of the tire pattern determination unit.

According to a seventh aspect of the present invention, a studless tire detection device includes the following image input unit, image storage unit, tire position detection unit, determination area extraction unit, tire pattern determination unit, and information presentation unit. The image input unit changes a shooting direction based on given position information,
Image input means for inputting an image captured by changing the zoom and focus based on the given position information as an input image, and an illumination for changing a light irradiation direction based on the given position information. I have. The image storage unit stores the input image. The tire position detector is means for detecting a tire position of the vehicle, sending the position information indicating the tire position to the image input unit, and generating a detection signal indicating that the tire position has been detected. The determination region extraction unit is a unit that receives the detection signal and extracts a region image of a predetermined size including the tire from the input image stored in the image storage unit. The tire pattern determination unit analyzes the extracted area image, extracts fine grooves on the surface of the tire as characteristics of the studless tire, and determines a studless tire mounting state of the vehicle based on the characteristic. is there. The information presentation unit is for performing the information presentation based on the determination result of the tire pattern determination unit.

According to an eighth aspect of the present invention, in the studless tire detection device according to the sixth or seventh aspect, the determination area extracting unit detects a ground contact surface of the tire from a luminance hiss drum of the input image, and Based on the input image, the rectangular region image of the tire portion is extracted. In a ninth aspect, in the studless tire detection device according to the eighth aspect, the determination region extracting unit includes:
A rectangle having the same size as the area image is moved along the ground plane, a position in the rectangle where the luminance or the luminance variance is minimum is detected, and the rectangle at this position is extracted as the area image. ing.

In a tenth aspect, in the studless tire detection device according to the sixth or seventh aspect, the determination area extracting section is configured to extract a rectangle at a predetermined position of the input image as the area image. ing. An eleventh invention is the studless tire detection device according to the sixth to ninth or tenth inventions, wherein the tire pattern determination unit comprises:
A binarization processing unit that binarizes the area image to obtain a binary image, an edge detection unit that extracts an edge in the binary image to obtain an edge image, a white pixel of the binary image,
An edge density calculation unit that determines an edge density of the tire from white pixels of the edge image, and a determination unit that determines whether the tire is a studless tire based on the edge density.

A twelfth aspect of the present invention is the studless tire detecting device according to the sixth to ninth or tenth aspects, wherein the tire pattern determination section includes a division processing section for dividing the area image into a plurality of pieces to obtain a small area image. A binarization processing unit that selects a threshold value from the luminance of each of the small region images, binarizes each of the small region images using the threshold value to obtain a plurality of binary images, An edge detection unit for extracting a plurality of edge images by extracting the edges in
White pixels of the plurality of edge images as a whole, an edge density calculation unit that calculates an edge density in the tire from white pixels of the region image, and a determination unit that determines whether the tire is a studless tire based on the edge density. It consists of:

According to a thirteenth aspect, in the studless tire detecting device according to the sixth to ninth or tenth aspects, the tire pattern determining section binarizes the area image to obtain a binary image. And an edge detecting unit for extracting an edge in the binary image to obtain an edge image, and replacing a small number of black pixels in a region composed of substantially white pixels in the binary image with white pixels to obtain a white image. A block separation unit that separates the image into blocks of pixel groups, a mask processing unit that extracts an image separated into the blocks as a mask image, and extracts a portion of the block from the region image, and a block processing unit that extracts the portion of the extracted block. An edge detection unit that detects the edge in the region image to obtain an edge detection image; a white pixel of the image divided into the blocks and an edge obtained by the edge detection unit. In the white pixel of the detected image and the edge density calculation unit for calculating an edge density in the tire, the tire from the edge density is the judging section that judges whether a studless tire, constitutes.

In a fourteenth aspect, in the studless tire detection device according to the twelfth to thirteenth aspects, the binarization processing section obtains a luminance histogram of a region to be processed.
The position of one peak value is detected, and the minimum value between the peak values is used as the threshold for binarization. 6th to 14th
According to the invention, as described above, the studless tire detection device is configured, so that the input image including the tire is obtained by the image input unit, and the region image is extracted from the input image by the determination region extraction unit. You. A fine groove on the tire surface in the region image is extracted by the tire pattern determination unit as a feature of the studless tire, and further, the mounting state of the studless tire is determined. Then, the information presentation unit presents information based on the determination result of the tire pattern determination unit.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS First Embodiment FIG. 1 is a configuration diagram of a tire chain detecting device according to a first embodiment of the present invention. This tire chain detecting device is a device for detecting a vehicle on which a tire chain is mounted on behalf of a person, and includes an image input means such as a CCD camera 11a, an illumination 11b, and an analog-digital converter (not shown). It has an input unit 11 and an image storage unit 12 for storing an input image Ig input from the image input unit 11. The image storage unit 12 includes a determination region extraction unit 13
Is connected to the output side of the determination area extracting unit 13. The output side of the chain attachment determination unit 14 is connected to the information presentation unit 15. The electrical signal S16 from the tire position detector 16 is input to the determination area extractor 13.

FIG. 2 is a block diagram showing the judgment area extracting unit 13 in FIG. The determination area extraction unit 13 includes the image storage unit 1
A region image R13 suitable for determining the mounting state of the tire chain is extracted from the image Ig stored in the image input unit 2. The image Ig is input from the image input unit 12 and a signal from the tire position detection unit 16 is input. A horizontal luminance histogram generator 13a to which S16 is input, a tire contact surface search unit 13b connected to the output side of the horizontal luminance histogram generator 13a, and a tire location search connected to the tire contact surface search unit 13c. The configuration is such that the region image R13 is output from the tire region exploration unit 13c.

FIG. 3 is a diagram showing the chain attachment determining unit 14 shown in FIG.
FIG. The chain mounting determination unit 14 analyzes the region image R13 given from the determination region extraction unit 13 to determine the mounting state of the tire chain, and includes a binarization processing unit 14a that binarizes the region R13. , A histogram generation unit 14b connected to the output side of the binarization processing unit 14a, and a histogram analysis unit 14c connected to the output side of the histogram generation unit 14b. The determination result J14 is output from the histogram analysis unit 14c.

The information presenting unit 15 includes the chain mounting determining unit 1
The warning information is presented to a supervisor or a driver on the basis of the determination result J14 given from No. 4, and various forms are conceivable as a presentation method. For example, there may be a case where a monitor simply shows whether or not a chain is mounted on a display or an alarm, or a case where a circuit breaker or the like is moved up and down to indicate whether or not a person is allowed to pass. The tire position detection unit 16 is configured by, for example, a photo interrupter switch or a tread plate using an infrared sensor, detects that the tire is passing through a detectable area of the device, and indicates a detection signal by an electric signal S16. It has become.

FIG. 4 is an explanatory view of a tire chain mounting inspection site. The tire chain detection device having the above-described configuration is set at a tire chain installation inspection site. The tire chain installation inspection site includes a parking lot and a chain attachment / detachment site, and the tire chain detection device is disposed along a guideway to the main line. Here, for example, due to the main road regulation
When prompting the vehicle 20 to install a tire chain, the vehicle 20 with the tire chain attached is guided to the main line, and the vehicle 20 without the tire chain is guided again to a parking lot or a chain attaching / detaching place. Conversely, when prompting the user to remove the tire chain at the entrance of a long tunnel during snowfall, etc., the vehicle 20 without the tire chain is guided to the main line, and the vehicle 20 with the tire chain is returned to the parking lot or the like. Guide to the chain attachment / detachment place.

FIGS. 5A and 5B are views showing an example of the arrangement of the image input unit 11 and the tire position detecting unit 16, wherein FIG. 5A is a plan view and FIG. FIG. For example, the CCD camera 11a and the illumination 11b constituting the image input unit 11 are arranged so as to photograph and irradiate the tire 21 of the vehicle 20 passing through the taxiway from diagonally forward or diagonally rear. In addition, the tire position detection unit 16
1 is arranged at a predetermined position in order to detect that it has come to a position where the CCD camera 11a can take a picture.

Next, the operation of the tire chain detecting device will be described. The illumination 11b irradiates the guideway with light obliquely, and the CCD camera 11a performs photographing. Image input section is C
An image photographed by the CD camera 11a is converted into an input image Ig composed of digital signals of multi-level gradation, and is sequentially provided to the image storage unit 12. The image storage unit 12 stores the input image Ig while updating it.

When the vehicle 20 that has entered the tire chain mounting inspection site crosses the infrared sensor of the tire position detector 16 or steps on the tread, the tire position detector 16 detects that the tire 21 has come to the predetermined position. Detect and signal S1
6 is sent to the determination area extraction unit 13. Judgment area extraction unit 13
Extracts the region image R13 as follows, given the signal S16. First, the horizontal luminance histogram generation unit 13a in the determination area extraction unit 13
From 2, the input image Ig at the time when the signal S16 is given is read.

FIGS. 6A and 6B are diagrams showing an input image Ig, and FIG. 7 is an explanatory diagram of a horizontal luminance histogram. When the vehicle 20 is equipped with a tire chain, the input image Ig read from the image storage unit 12 includes a tire 21 and a tire chain 22 attached thereto, as shown in FIG. In the case of the vehicle 20 without a tire chain, the tire chain 22 does not exist. The horizontal direction histogram creation unit 13a in the determination area extraction unit 13a creates a histogram S13a by adding the luminance of the input image Ig in the horizontal direction. When the histogram S13a of the input image Ig in FIG. 6A is created, as shown in FIG. 7, the histogram value of the road surface portion is high, and the portion corresponding to the tire 21 above the road surface is low and substantially constant. The histogram S13a is the tire contact surface search unit 1
3b. The tire contact surface searching unit 13b observes the histogram S13a from below, and obtains a position where the histogram value of the luminance sharply decreases and becomes constant as the contact surface 23 between the tire 21 and the road surface. The information on the contact surface 23 is provided to the tire portion search unit 13c.

FIG. 8 is an explanatory diagram of a tire site search.
The tire part exploring unit 13c sets a rectangular area 24 having the same size as the area image R13 in the image Ig, and
Is moved in accordance with the contact surface 23, and its luminance distribution is observed. As a result of the observation, the region R13 corresponding to the rectangular region 24 at the position where the variance of the luminance distribution is minimum is determined as the tire part. By this processing, the tread portion of the tire 21 is extracted as the region image R13. The region image R13 is
It is sent to the chain attachment determination unit 14.

FIG. 9 is a diagram showing a region image R13 cut out from the input image Ig, and FIG. 10 is a diagram showing a result of the binarization process of FIG. When the vehicle 20 is wearing a tire chain, the region image R13 extracted and cut out from the input image Ig includes the tread of the tire 21 and the tire chain 22. The binarization processing unit 14a in the chain attachment determination unit 14 that has received the region image R13 performs binarization on the region image R13 using a threshold. Here, if the threshold is set appropriately, the binarization processing result S
14a, as shown in FIG.
And the shadow 25 of the chain 22 remain.

FIG. 11 is a diagram showing the histogram of FIG. Binarization processing result S14 from binarization processing unit 14a
a, the histogram generation unit 14b having received the horizontal luminance in the binarization processing result S14a adds
A histogram S14b as shown in FIG. The histogram analysis unit 14c calculates the histogram S14b
A portion 26 having a small peak value is extracted, and is extracted as 1
If there are more than two locations, it is determined that the vehicle 20 is wearing a tire chain. The determination result J14 is provided to the information presenting unit 15, and is notified by moving up or down the circuit breaker or displaying the same to a supervisor.

As described above, in the first embodiment,
The image input unit 11, the image storage unit 12, the determination region extraction unit 13, the chain attachment determination unit 14, and the information presentation unit 15 constitute a tire chain detection device, and the input image Ig input by the image input unit 11 Therefore, since the configuration is such that it is determined whether or not the vehicle 20 is equipped with a tire chain, it is not necessary for a human to work at night or in a place where working conditions are bad such as bad weather. In addition, since it can be determined only by basic image processing techniques such as binarization and histogram creation,
The configuration of the entire apparatus can be made simple and inexpensive. Further, since the determination of the attachment of the chain 22 uses the shadow 25 of the light applied by the illumination 11b, a stable determination result can be obtained without being affected by the material of the chain 22.

Second Embodiment FIG. 12 is a configuration diagram of a tire chain detecting device according to a second embodiment of the present invention. This tire chain detection device is a device that can determine whether or not the vehicle 20 is wearing the tire chain 22 even when the passing vehicle 20 does not pass through a predetermined position at the tire chain installation inspection site, An image input unit 31 different from the first embodiment, an image storage unit 32 similar to the first embodiment,
A determination area extraction unit 33, a chain attachment determination unit 34, an information presentation unit 35, and a tire position detection unit 36 different from the first embodiment are provided.

The image input unit 31 is a CCD camera 31a having a function of changing the focus and zoom as image input means.
And the illumination 31b. The CCD camera 31a and the illumination 31b are attached to a camera platform 31c that electrically changes the photographing direction and the light irradiation direction of the CCD camera 31a and the illumination 31b. The image input unit 31 has an analog-to-digital converter (not shown). On the output side of the image input unit 31, an image storage unit 32, a determination region extraction unit 33, a chain attachment determination unit 34, and an information presentation unit 35 are sequentially connected. The tire position detection unit 36 detects the tire position of the vehicle 20 that has entered the taxiway, and provides an electric signal S36a of position information indicating the tire position to the platform 31c and the CCD camera 31a, and detects the tire position. Detection signal S36b indicating that the
3, for example, a plurality of infrared sensors or treads disposed at a plurality of locations and a signal S36a,
It is composed of a control circuit for generating S36b. The tire chain detection device having the above-described configuration is installed in a tire chain installation inspection site as shown in FIG. 4, as in the first embodiment.

FIGS. 13 (a) and 13 (b) are views showing examples of the arrangement of the image input unit 31 and the tire position detecting unit 36 in FIG. 12, wherein FIG. 13 (a) is a plan view and FIG. ) Is a side view. For example, the CCD camera 3 constituting the image input unit 31
1a and the lighting 31b are provided for the vehicle 20 that normally passes through the taxiway.
The tire 21 is arranged so as to shoot and irradiate the tire 21 from obliquely forward or obliquely backward. Also, the tire position detection unit 3
In the case where a plurality of infrared sensors are used in 6, the plurality of infrared sensors are arranged along the taxiway to detect that the infrared sensors have come to a position where the CCD camera 31a can capture an image.

Next, the operation of the tire chain detecting device will be described. The illumination 31b irradiates light in the direction determined by the camera platform 31c, and the CCD camera 31a captures an image. An image photographed by the CCD camera 31a is converted into an input image Ig composed of digital signals of multi-level gradation, and is sequentially provided to the image storage unit 32. Image storage unit 32
Stores the input image Ig while updating it. Here, when the intruding vehicle 20 crosses the plurality of infrared sensors of the tire position detection unit 36 or steps on the tread plate, the tire position detection unit 36 determines, based on the position of the crossed infrared sensor and the reflection from the tire 21, Alternatively, from the position of the tread, the tire 21
Is detected, and an electric signal S36 corresponding to the position information is detected.
a is generated and provided to the camera platform 31c and the CCD camera 31a, and a detection signal S36b indicating the detection is generated and provided to the determination area extraction unit 33.

Head 31c to which electric signal S36a is input
Turns the CCD camera 31a and the illumination 31b toward the tire 21. The CCD camera 31a adjusts the zoom and focus to the position of the tire 21. Thereby, the input image I
In g, the tire 21 of the vehicle 20 is shown. This input image Ig is stored in the image storage unit 32. Detection signal S36b
Is read from the image storage unit 32 at the time when the signal S36b is supplied, and extracts a region image R33 corresponding to the region image R13 in the first embodiment. In this case, since the input image Ig has already been included only in the tire 21, only a predetermined area of the input image Ig may be cut out.

The region image R33 is stored in the chain
4 As in the first embodiment, the chain attachment determination unit 34 binarizes the region image R33, obtains a histogram thereof, and analyzes the histogram to determine the vehicle 20
It is determined whether the tire chain 22 is mounted or not mounted in. The determination result J34 is given to the information presenting unit 35, and is notified by moving up or down the circuit breaker or displaying it to a monitoring person.

As described above, in the second embodiment,
The image input unit 31, the image storage unit 32, the determination region extraction unit 33, the chain attachment determination unit 34, and the information presentation unit 35 constitute a tire chain detection device, and the input image Ig input by the image input unit 31 Therefore, since the configuration is such that it is determined whether or not the vehicle 20 is equipped with a tire chain, it is not necessary for a human to work at night or in a place where working conditions are bad such as bad weather. In addition, since the determination can be made only by basic image processing techniques such as binarization and creation of a histogram, the configuration of the entire apparatus can be made simple and inexpensive.
A stable determination result can be obtained without being affected by the material of the chain 22 in determining the mounting. Further, the tire position is detected by the tire position detection unit 36 and the electric signal S3 is detected.
Since the pan head 31c is notified at 6a and the position of the tire 21 is automatically photographed by the CCD camera 31a, restrictions on the arrangement accuracy, the driver's skill, and the like can be relaxed.

Third Embodiment FIG. 14 is a configuration diagram of a studless tire detection device according to a third embodiment of the present invention. This studless tire detection device is used for a vehicle 2 equipped with studless tires.
This is a device for detecting 0 on behalf of a person.
a image input unit 41 including an image input unit such as a, an illumination 41b, an analog-to-digital converter (not shown), and an image storage unit 42 for storing an image Ig input from the image input unit 41. I have. The image storage unit 42 is connected to a determination region extraction unit 43, and an output side of the determination region extraction unit 43 is connected to a tire pattern determination unit 44.
The output side of the tire pattern determination unit 44 is
It is connected to the. The determination area extraction unit 43 is configured to receive a signal S46 from the tire position detection unit 46.

FIG. 15 is a diagram showing the judgment area extracting unit 43 in FIG.
FIG. The determination area extraction unit 43, as in the first embodiment, stores the image Ig stored in the image storage unit 42.
The region image R43 suitable for judging the wearing state of the studless tire is extracted from the horizontal direction. The image Ig is input from the image input unit 42 and the signal S46 from the tire position detection unit 46 is input. It has a direction luminance histogram generation unit 43a, a tire contact surface search unit 43b connected to the output side of the horizontal direction luminance histogram generation unit 43a, and a tire part search unit 43c connected to the tire contact surface search unit 43b. are doing. The configuration is such that the region image R43 is output from the tire portion search unit 43c.

FIG. 16 is a block diagram showing the tire pattern judging section 44 in FIG. Tire pattern determination unit 44
Is the region image R43 given from the determination region extraction unit 43.
And a binarization processing section 44a for binarizing the region image R43.
An edge detection unit 44b, an edge density calculation unit 44c, and a determination unit 44d are sequentially connected to the output side of the value processing unit 44a. The edge detection unit 44b includes a binarization processing unit 4
It has a function of performing edge detection on the binary image S44a given from 4a and outputting an edge image S44b. The edge density calculator 44c calculates the edge density S44c based on the binary image S44a and the edge image S44b, and gives the calculated edge density S44c to the determiner 44d. Judgment unit 44d
Has a function of determining whether or not the vehicle 20 is wearing studless tires from the edge density S44c, and the determining unit 44d outputs a determination result J44 to the information presenting unit 45.

The information presenting section 45 presents warning information to a supervisor or a driver based on the judgment result J44 given from the tire pattern judging section 44. The presenting method has various forms. Conceivable. For example, there may be a case where the monitor simply indicates whether or not a studless tire is being worn, and a case where it is possible to raise or lower a circuit breaker or the like to indicate whether or not traffic is possible. The tire position detection unit 46 is configured by, for example, a photo interrupter switch or a tread plate using an infrared sensor, detects that the tire passes through the detectable area of the device, and has a configuration indicated by an electric signal S46. I have.

FIG. 17 is an explanatory view of a studless tire mounting inspection site. The studless tire detection device having the above-described configuration is set in a studless tire mounting inspection site. The studless tire detection device is disposed along a taxiway from the parking lot to the main line. Here, for example, in order to prompt the vehicle 20 heading to the main line to use a non-slip device or the like due to road regulation on the main line, the vehicle 20 with studless tires is moved to the main line, and the vehicle 20 without studless tires is moved again to the parking lot. Induce.

FIGS. 18A and 18B show the image input section 41.
4A and 4B are diagrams illustrating an example of the arrangement of the tire position detection unit 46, wherein FIG. 4A is a plan view and FIG. 4B is a side view. For example, the CCD camera 41a and the illumination 41b constituting the image input unit 41 shoot and illuminate the tire 21 of the vehicle 20 passing through the taxiway from, for example, obliquely forward or obliquely rearward, as in the first embodiment. Placed in Further, the tire position detecting unit 46 detects that the tire 21
It is arranged at a predetermined position in order to detect that it has come to a position where a photograph can be taken at a.

Next, the operation of the tire chain detecting device will be described. The illumination 41b irradiates the guideway with light obliquely, and the CCD camera 41a performs photographing. Image input unit 41
Converts an image photographed by the CCD camera 41a into an input image Ig composed of digital signals of multi-level gradation, and sequentially supplies the input image Ig to the image storage unit 42. The image storage unit 42 stores the input image I
Store g while updating it. When the vehicle 20 that has entered the studless tire mounting inspection site crosses the infrared sensor of the tire position detection unit 46 or steps on the tread, the tire position detection unit 46 detects that the tire 21 has come to its predetermined position, The detection signal S46 is sent to the determination area extraction unit 43.

When given the signal S46, the judgment area extracting section 43 extracts the area image R43 as follows.
The horizontal direction luminance histogram generation unit 43a in the determination region extraction unit 43 reads the input image Ig at the time when the signal S46 is given from the image storage unit 42. The input image Ig includes the tire 21 as shown in FIG.
Image is shown. Horizontal histogram generator 43
In a, similarly to the first embodiment, the luminance of the input image Ig is added in the horizontal direction to create a histogram S43a.
In the histogram S43a, the histogram value of the road surface portion is high, and the portion corresponding to the tire 21 above the road surface is low and substantially constant. The histogram S43a is provided to the tire contact surface searching unit 43b. As in the first embodiment, the tire ground contact surface search unit 43b outputs the histogram S4
3a is observed from below, and the position where the histogram value of the luminance decreases rapidly and becomes constant is the contact surface 2 between the tire 21 and the road surface.
Calculate as 3. The information S43b of the contact surface 23 is given to the tire portion search unit 43c.

The tire part exploring section 43c outputs the area image R4
A rectangular area having the same size as 3 is set in the image Ig, and the rectangular area is moved in accordance with the contact surface, and its luminance distribution is observed. As a result of the observation, the region image R43 corresponding to the rectangular region at the position where the variance of the luminance distribution is minimized is determined as the tire region. By this processing, the tread portion of the tire 21 is extracted as the region image R43. The region image R43 is
The tire pattern is sent to the tire pattern determination unit 44.

FIG. 19 is a diagram showing a binary image S44a corresponding to the region image R43. Binary processing section 44a in tire pattern determination section 44 to which region image R43 has been input
Creates the brightness histogram of the brightness versus the frequency by first obtaining the brightness of the pixels of the entire region image R43. In this luminance histogram, two peaks composed of the luminance of the groove of the tire 21 and the luminance of the surface of the tire 21 appear. The binarization processing unit 44a binarizes the region image R43 by selecting a luminance indicating the minimum frequency sandwiched between these two peaks as a threshold. As a result of the binarization, for example, as shown in FIG. 19, a binary image S44a showing the groove and the surface of the tire 21 is obtained. The binary image S44a is provided to the edge detection unit 44b.

FIG. 20 is a diagram showing an edge image S44b obtained from the binary image S44a. The edge detection unit 44b that has received the binary image S44a detects an edge of the binary image S44a. For this processing, Laplacian filtering processing or the like generally used in image processing is used. By this filtering process, a grayscale image is obtained in which a large value is assigned to an edge portion and a small value is assigned to a flat portion. The edge detection unit 44 applies an appropriate threshold value to the grayscale image to binarize the image, and the edge image S
This is given to the edge density calculation unit 44c as 44b.

The edge density calculation unit 44c calculates the number N of white pixels in the binary image S44a obtained by the binarization processing unit 44a and the number M of white pixels in the edge image S44b, and calculates the ratio (M / N) is calculated as the edge density, and this is given to the determination unit 44d. The determination unit 44d calculates the ratio (M / N)
Is compared with a preset threshold T. If the ratio (M / N) is larger than the threshold T, the tire 21 is determined to be a studless tire, and the ratio (M / N) is larger than the threshold T. Is smaller than the normal tire. Such a determination is a determination based on the characteristic that a small groove is generally formed on the tire surface in a studless tire. The determination result J44 is provided to the information presenting unit 45, and is notified by raising or lowering the circuit breaker or displaying it to a supervisor.

As described above, in the third embodiment,
An image input unit 41, an image storage unit 42, a determination area extraction unit 43, a tire pattern determination unit 44, and an information presentation unit 45
And a tire position detector 46 constitute a studless tire detector, and the input image I input by the image input unit 41
Since the vehicle 20 equipped with studless tires and the vehicle 20 equipped with normal tires are determined from g, humans do not need to work at night or in places with poor working conditions such as bad weather. Also, using only basic image processing techniques such as binarization and histogram creation,
Since the determination can be made, the configuration of the entire apparatus can be made simple and inexpensive, and the determination can be made at a speed and an inspection accuracy equal to or higher than the inspection performed by a human.

Fourth Embodiment FIG. 21 is a configuration diagram of a studless tire detection device according to a fourth embodiment of the present invention. This studless tire detection device is a device that can stably detect a vehicle 20 with studless tires even when the brightness of the tire surface changes continuously due to the influence of light applied to the tire surface. And an image input unit 51 similar to the third embodiment, which includes image input means such as a CCD camera 51a, illumination 51b, an analog-to-digital converter (not shown), and an image input from the image input unit 51. And an image storage unit 52 for storing Ig. Image storage unit 52
Is connected to a determination area extraction unit 53, and an output side of the determination area extraction unit 53 is connected to a tire pattern determination unit 54. The output side of the tire pattern determination unit 54 is connected to the information presentation unit 55. The determination area extraction unit 53 is configured to receive a signal S56 from the tire position detection unit 56. The image storage unit 52, the determination area extraction unit 53, the information presentation unit 55, and the tire position detection unit 56
The configuration is the same as that of the third embodiment, and the configuration of the tire pattern determination unit 54 is different from that of the third embodiment.

FIG. 22 is a configuration diagram showing the tire pattern determination section 54 in FIG. The tire pattern determination unit 54 determines the region image R given from the determination region extraction unit 53.
53 is analyzed to determine a tire pattern.
A division processing unit 54a that divides the region image R53, and a binarization processing unit 54b that binarizes each divided region image
And an edge detection unit 54c, an edge density calculation unit 54d, and a determination unit 54e that are sequentially connected to the output side of the binarization processing unit 54b. The edge detection unit 54c has a function of performing edge detection on the binary image S54b provided from the binarization processing unit 54b and outputting an edge image S54c. The edge density calculation unit 54d calculates the binary image S
54b and the edge density S5 based on the edge image S54c.
4d is calculated and given to the determination unit 54e. The determination unit 54e has a function of determining whether or not the vehicle 20 is wearing studless tires from the edge density S54d, and the determination unit 54e gives the information presentation unit 55 a determination result J54.
Is output.

Next, the operation of the tire chain detecting device will be described. The image input unit 51 and the image storage unit 52 operate in the same manner as in the third embodiment, and the input image Ig is stored in the image storage unit 52. When the vehicle 20 that has entered the studless tire mounting inspection site crosses the infrared sensor of the tire position detection unit 56 or steps on the tread, the tire position detection unit 56 detects that the tire 21 has come to its predetermined position, The signal S56 is sent to the determination area extraction unit 53. When the signal S56 is given, the determination area extraction unit 53 extracts the area image R53 by the same processing as in the third embodiment.

FIG. 23 is a diagram showing division of the region image R53. The division processing unit 54a in the tire pattern determination unit 54 that has received the region image R53 divides the region image R53 into a plurality of small region images S54a. The number of divisions can be set arbitrarily, but it is better to divide the divisions evenly for the execution of post-processing. Further, the size of the small area image S54a can be freely set. The plurality of small area images S54a are provided to the binarization processing unit 54b.

FIG. 24 is an explanatory diagram of the binarization processing of the small area image S54a. The binarization processing unit 54b obtains a histogram of luminance versus frequency for each small area image S54a. In each histogram, two peaks appear from the characteristics of the tire surface. The binarization processing unit 54b binarizes each of the small area images S54a by using the minimum value between the peaks of each of the small area images S54a as a threshold for binarization. In this way, by performing binarization for each of the plurality of small area images S54a, it is possible to perform binarization that does not receive a change in luminance over the entire area image R53. The binarization processing result S54b for all the small area images S54a is given to the edge detection unit 54c. Edge detector 54c
Makes the binarization processing result S54b into one image, detects an edge from the binarization processing result S54b, generates an edge image S54c, and gives the edge image S54c to the edge density calculation unit 54d.
The edge density calculation unit 54d and the determination unit 54e determine whether the tire 21 is a studless tire or a normal tire from the edge image S54c by the same processing as in the third embodiment, and provides the information presentation unit 55 with the determination result J54. The information presenting unit 55 also presents information to a supervisor or a driver, as in the third embodiment.

As described above, in the fourth embodiment,
Since the division processing unit 54a is provided in the tire pattern determination unit 54, in addition to the same effects as in the third embodiment, the tire 21
Even if the brightness of the tire surface changes continuously depending on how the light hits the area, stable binarization of the region image R53 can be performed, so that the mounted state of the studless tire can be determined more accurately.

Fifth Embodiment FIG. 25 is a configuration diagram of a studless ear detection device according to a fifth embodiment of the present invention. This studless tire detection device is a device that determines whether or not a studless tire is worn even when a fine groove on the tire surface that is a feature of the studless tire disappears due to binarization. The image input unit 61 includes an image input unit 61 similar to the third embodiment including an analog-to-digital converter 61b and an analog / digital converter (not shown), and an image storage unit 62 that stores an image Ig input from the image input unit 61. I have. The image storage unit 62 is connected to a determination region extraction unit 63, and an output side of the determination region extraction unit 63 is connected to a tire pattern determination unit 64. The output side of the tire pattern determination unit 64 is connected to the information presentation unit 65. The determination area extraction unit 63 is configured to receive a signal S66 from the tire position detection unit 66. Image storage unit 6
2. The determination area extraction unit 63, the information presentation unit 65, and the tire position detection unit 66 have the same configuration as that of the third embodiment, and the configuration of the tire pattern determination unit 64 is different from that of the third embodiment. .

FIG. 26 is a configuration diagram showing the tire pattern determination section 64 in FIG. The tire pattern determination unit 64 determines the region image R given from the determination region extraction unit 63.
63 is analyzed to determine a tire pattern.
A binarization processing unit 64a similar to the third embodiment for binarizing the region image R63, a block separation unit 64b sequentially connected to the output side of the binarization processing unit 64a, and a mask processing unit 64c, an edge detection unit 64d, an edge density calculation unit 64e, and a determination unit 64f.

The block separating section 64b includes a binarizing section 6
From the binary image S64a given from 4a, a block of a convex portion separated by a large groove on the tire surface is separated. The mask processing unit 64c sets the block image S64b as a mask image and uses the region image R6 as an original image.
This is a means for extracting the block portion S64d from the block 3. The edge detection unit 64d, the edge density calculation unit 64e, and the determination unit 64f have functions similar to those of the third embodiment.

Next, the operation of the studless tire detecting device will be described. Image input unit 61 and image storage unit 62
Operates in the same manner as the third embodiment, and the input image Ig is
It is stored in the image storage unit 62. When the vehicle 20 that has entered the studless tire mounting inspection site crosses the infrared sensor of the tire position detection unit 66 or steps on the tread, the tire position detection unit 66 detects that the tire 21 has come to its predetermined position, The detection signal S66 is sent to the determination area extraction unit 63. When given the signal S66, the determination area extraction unit 63 extracts the area image R63 by the same processing as in the third embodiment. The binarization processing unit 64a of the tire pattern determination unit 64 to which the region image R63 has been input generates and outputs a binary image S64a of the region image R63 by the same processing as in the third embodiment.

FIG. 27 is an explanatory diagram of the block separation processing. The block separation unit 64b determines the given binary image S6
4a is subjected to a filling process to separate into blocks BL. That is, in the binary image S64a, the tire surface is white,
Since the deep groove is black, processing for filling a small black hole in the white area is performed as follows. First, labeling processing is performed on black pixels of the entire binary image S64a, and independent regions are numbered. If the area (number of pixels) of each independent region is less than a certain size, the region is filled with white pixels. Such processing is performed on the binary image S64a.
By performing the entire process, fine grooves and holes in the block BL are closed, and the blocks BL can be separated into blocks. The block image S64b formed by the separation of the block BL is provided to the mask processing unit 64c and the edge density calculation unit 64e.

The mask processing section 64c outputs the block image S6
Using 4b as a mask image, a portion corresponding to the block BL is extracted from the region image R63. Specifically, a portion corresponding to a white region of the mask image is extracted from the region image R63, and a black pixel portion is left as it is. The edge detection unit 64e outputs the image S obtained from the mask processing unit S64c.
Edge detection is performed on 64c in the same manner as in the third embodiment, and an edge image S64d of the detection result is provided to the edge density calculation unit 64e. Here, even if the fine grooves on the tire surface of the studless tire disappear in the binarization process,
A region image R6 which is an original image of a portion where block separation has been performed.
Since edge detection is performed based on No. 3, not only the edge of the block itself but also the edge of a fine groove on the tire surface can be detected. Furthermore, since the groove portion between the blocks is filled with black pixels, no edge is detected for a change in luminance existing in the groove, so that no extra edge is detected as compared with the case where the original image is directly detected as an edge. become.

Edge density calculation section 64e and determination section 64f
Determines whether the tire 21 is a studless tire or a normal tire by the same processing as in the third embodiment, and gives the determination result J64 to the information presenting unit 65. Information presentation unit 65
Indicates to the supervisor or the driver whether the vehicle 20 is wearing studless tires or normal tires, as in the third embodiment.

As described above, in the fifth embodiment,
Since the tire pattern determination unit 64 is provided with the block separation unit 64b and the mask processing unit 64c, in addition to the same effect as in the third embodiment, the case where fine grooves on the tire surface of a studless tire disappear by binarization processing However, since edge extraction is performed based on the region image R63 of the portion where the block is separated, not only the edge of the block itself but also the edge of a fine groove on the tire surface can be detected, so that a studless tire can be accurately determined, The detection accuracy of the vehicle 20 equipped with the tire is improved.

Sixth Embodiment FIG. 28 is a configuration diagram of a studless tire detection device according to a sixth embodiment of the present invention. This studless tire detection device, even when the vehicle 20 does not pass through a predetermined position at a studless tire mounting inspection site,
This is a device that can determine whether or not the vehicle 20 is wearing studless tires. The image input unit 71 is different from the third embodiment, and the image storage unit 7 is the same as that of the third embodiment.
2, a determination area extracting unit 73, a tire pattern determining unit 74, an information presenting unit 75, and a tire position detecting unit 76 different from the third embodiment.

The image input unit 71 is a CCD camera 71a having a function of changing the focus and zoom as image input means.
And the illumination 71b. The CCD camera 71a and the illumination 71b are attached to a camera platform 71c that electrically changes the shooting direction and the light irradiation direction of the CCD camera 71a and the illumination 71b. The image input unit 71 has an analog-to-digital converter (not shown). On the output side of the image input unit 71, an image storage unit 72, a determination area extraction unit 73, a tire pattern determination unit 74, and an information presentation unit 75
And are connected in order. The tire position detector 76 detects the tire position of the invading vehicle 20 and provides an electric signal S76a as position information to the platform 71c, and extracts a detection signal S76b indicating that the tire position has been detected as a determination area. It has a function to be provided to the unit 73, and includes, for example, a plurality of infrared sensors or treads arranged at a plurality of locations, and a control circuit for generating the signals S76a and S76b. Image input unit 71 and tire position detection unit 76
Are arranged on the taxiway of the studless tire inspection site in the same manner as in the second embodiment.

Next, the operation of the studless tire detection device will be described. The illumination 71b irradiates light in the direction determined by the camera platform 71c, and the CCD camera 71a captures an image. The image captured by the CCD camera 71a is converted into an input image Ig composed of digital signals of multi-valued gradation, and is sequentially provided to the image storage unit 72. Image storage unit 7
2 stores the input image Ig while updating it. Here, when the vehicle 20 that has entered the studless tire mounting inspection site crosses or steps on a plurality of infrared sensors of the tire position detection unit 76, the tire position detection unit 76 determines the position of the crossed infrared sensor and the tire position. The position of the tire 21 is detected from the reflection from the position 21 or the position of the tread plate, and an electric signal S76a, which is the position information, is generated and supplied to the camera platform 71c and the CCD camera 71a. The signal S76b is generated and provided to the determination area extraction unit 73.

The platform 31c to which the electric signal S76a is input
Turns the CCD camera 71a and the illumination 71b toward the tire 21. The CCD camera 71a adjusts the zoom and focus to the position of the tire 21. Thereby, the input image Ig
Contains the tire 21 of the vehicle 20 and the input image I
g is stored in the image storage unit 72. Upon receiving the detection signal S76b, the determination region extraction unit 73 reads the input image Ig at the time when the signal S76b is supplied from the image storage unit 72, and extracts a region image R73 corresponding to the region image R63 in the third embodiment. I do. In this case, since only the tire 21 is already included in the input image Ig, a predetermined area of the input image Ig may be simply cut out.

The area image R 73 is sent to the tire pattern judgment section 74. The tire pattern determination unit 74 determines the third,
Similarly to the fourth or fifth embodiment, binarization of the region image R73 and edge detection are performed, and it is determined whether the tire 21 is a studless tire or a normal tire based on the number of white pixels of the binary image and the edge image. Then, the determination result J74 is output to the information presentation unit 75. The information presenting unit 75 determines the determination result J7
Based on 4 above, it raises and lowers the circuit breaker, or notifies the observer by display or the like.

As described above, in the sixth embodiment,
An image input unit 71, an image storage unit 72, a determination area extraction unit 73, a tire pattern determination unit 74, and an information presentation unit 75
And a tire position detector 76. The tire position detector 76 detects the tire position, notifies the camera platform 71c and the CCD camera 71a with a signal S76a, and automatically captures the position of the tire 21 with the CCD camera 71a. Therefore, restrictions on placement accuracy, driver skill, and the like can be relaxed.

The present invention is not limited to the above embodiment, but can be variously modified. For example, there are the following modifications. (1) In the first, third to fifth embodiments, the image storage units 12, 42, 52, and 62 sequentially store the input image Ig,
Although the determination area extracting means 13, 43, 53, 63 extracts the area images R13, R43, R53, R63 in synchronization with the detection signal, the image storage units 12, 42, 52,
62 may store the input image Ig in synchronization with the detection signal. (2) The determination area extracting means 13, 43, 53, 63
Since it is sufficient that the image data of the tire part can be extracted, if the input image Ig can capture the same part of the tire 21 at the same position, a rectangle at a fixed position may be extracted as an area image. This eliminates the need for comparison for extracting a rectangular area image, and can speed up the processing.

(3) The configuration of the chain attachment determining units 14 and 24 is only required to be able to extract the shadow of the chain 22, and is not limited to the configurations of the first and second embodiments. For example, even if edge detection is performed only in the horizontal direction of the region images R13 and R23, a result responsive to the shadow is obtained.
The same effects as in the above embodiment can be obtained. (4) Image input unit 5 in the fourth and sixth embodiments
1, 61 and the tire position detectors 56 and 66 can be replaced by the image input unit 71 and the tire position detector 76 of the sixth embodiment. (5) CCD cameras 31a, 71 serving as image input means
The a and the illuminations 31b and 71b may have a function of changing a shooting direction and a light irradiation direction by the CCD camera itself and the illumination itself. In this case, the CCD cameras 31a, 7
Head 31 for installing 1a and lights 31b, 71b
c and 71c may not have the function of changing the direction.

[0070]

As described in detail above, the first to fifth embodiments
According to the invention, a tire chain detection device is configured by an image input unit, an image storage unit, a tire position detection unit, a determination region extraction unit, a chain attachment determination unit, and an information presentation unit, and an input image including a tire is input to the image input unit. The area image is extracted by the determination area extraction unit, and the state of the tire chain mounting of the vehicle is determined by analyzing the area image.
A vehicle equipped with a tire chain can be detected instead of a person. Therefore, it is possible to alleviate harsh working conditions and improve the inspection speed and inspection accuracy.

According to the sixth to fourteenth aspects, an image input unit, an image storage unit, a tire position detection unit, a determination region extraction unit,
A tire pattern determination unit and an information presentation unit constitute a studless tire detection device, an image input unit obtains an input image of a tire, and a determination region extraction unit extracts a region image from the input image to obtain a region image. Since the small grooves on the tire surface are extracted as characteristics of the studless tire and the mounting state of the studless tire is determined, a vehicle wearing the studless tire can be detected instead of a person. Therefore, it is possible to alleviate harsh working conditions and improve the inspection speed and inspection accuracy.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a tire chain detection device according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram showing a determination area extraction unit 13 in FIG.

FIG. 3 is a configuration diagram showing a chain attachment determination unit 14 in FIG. 1;

FIG. 4 is an explanatory diagram of a tire chain installation inspection site.

FIG. 5 is a diagram showing an example of the arrangement of an image input unit 10 and a tire position detection unit 16;

FIG. 6 is a diagram showing an input image Ig.

FIG. 7 is an explanatory diagram of a horizontal luminance histogram.

FIG. 8 is an explanatory diagram of a tire part search.

FIG. 9 is a diagram showing a region image R13 cut out from an input image Ig.

FIG. 10 is a diagram showing a result of the binarization processing of FIG. 9;

FIG. 11 is a diagram showing the histogram of FIG. 10;

FIG. 12 is a configuration diagram of a tire chain detection device according to a second embodiment of the present invention.

13 is a diagram illustrating an example of the arrangement of an image input unit 31 and a tire position detection unit 36 in FIG.

FIG. 14 is a configuration diagram of a studless tire detection device according to a third embodiment of the present invention.

FIG. 15 is a configuration diagram illustrating a determination area extraction unit 43 in FIG. 14;

16 is a configuration diagram showing a tire pattern determination unit 44 in FIG.

FIG. 17 is an explanatory diagram of a studless tire mounting inspection site.

FIG. 18 is a diagram illustrating an example of the arrangement of an image input unit 41 and a tire position detection unit 46.

FIG. 19 shows a binary image S44a corresponding to a region image R43.
FIG.

FIG. 20 is an edge image S4 obtained from a binary image S44a.
It is a figure showing 4b.

FIG. 21 is a configuration diagram of a studless tire detection device according to a fourth embodiment of the present invention.

FIG. 22 is a configuration diagram illustrating a tire pattern determination unit in FIG. 21;

FIG. 23 is a diagram showing division of a region image R53.

FIG. 24 is an explanatory diagram of the binarization processing of the small area image S54a.

FIG. 25 is a configuration diagram of a studless ear detection device according to a fifth embodiment of the present invention.

26 is a configuration diagram illustrating a tire pattern determination unit 64 in FIG.

FIG. 27 is an explanatory diagram of a block separation process.

FIG. 28 is a configuration diagram of a studless tire detection device according to a sixth embodiment of the present invention.

[Explanation of symbols]

11, 31, 41, 51, 61, 71 Image input units 11a, 31a, 41a, 51a, 61a, 71a
CCD camera 11b, 31b, 41b, 51b, 61b, 71b
Illumination 31c, 71c Pan head 12, 32, 42, 52, 62, 72 Image storage unit 13, 33, 43, 53, 63, 73 Judgment area extraction unit 14, 34 Chain attachment judgment unit 15, 35, 45, 55, 65, 75 Information presentation unit 16, 36, 46, 56, 66, 76 Tire position detection unit 44, 54, 64, 74 Studless tire determination unit Ig Input image R13, R33, R43, R53, R63, R73
Area image

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Naohiro Tenmoto 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd. F-term (reference) 2F065 AA02 AA12 AA67 CC00 CC13 DD06 FF01 FF02 FF04 FF67 HH12 HH15 HH18 JJ03 JJ05 JJ08 JJ26 PP01 QQ04 QQ24 QQ25 QQ26 QQ29 QQ33 QQ36 QQ43 SS09 5B057 AA16 BA15 BA30 CH11 DA02 DA08 DA12 DB02 DB08 DC16 DC23

Claims (14)

[Claims] 1. A tire chain detecting device for detecting a mounted state of a tire chain of a vehicle and presenting information of the detection result, comprising: an image inputting means for taking an image of a predetermined place as an input image; An image storage unit that stores the input image, and detects that the tire of the vehicle has arrived at the predetermined location, generates a detection signal, and indicates that the tire is captured in the input image. A tire position detection unit, a determination region extraction unit that receives the detection signal, and extracts a region image of a predetermined size including the tire from the input image stored in the image storage unit; Analyzing a region image to extract a shadow formed by the lighting, and determining a tire chain mounting state of the vehicle based on the shadow; a chain mounting determining unit; An information presenting unit for presenting the information based on a fixed result. 2. A tire chain detecting device that detects a tire chain mounting state of a vehicle and presents information of the detection result, wherein a photographing direction is changed based on given position information, and the given position information is changed. Image input means for inputting an image captured by changing the zoom and focus based on the input image, an image input unit having illumination for changing a light irradiation direction based on the given position information, and An image storage unit for storing, a tire position detecting unit for detecting a tire position of the vehicle, transmitting the position information indicating the tire position to the image input unit, and generating a detection signal indicating that the tire position is detected. A determination area extracting step of receiving the detection signal and extracting an area image of a predetermined size including the tire from the input image stored in the image storage unit. A chain mounting determination unit that analyzes the extracted region image to extract a shadow formed by the lighting, and determines a tire chain mounting state of the vehicle based on the shadow; and a determination by the chain mounting determination unit. An information presenting unit for presenting the information based on a result. 3. The determination area extraction unit detects a ground contact surface of the tire from a luminance histogram of the input image, and extracts the rectangular region image of the tire portion in the input image based on the contact surface. The tire chain detecting device according to claim 1 or 2, wherein the tire chain detecting device is configured. 4. The determination area extraction unit moves a rectangle having the same size as the area image along the ground plane, and detects a position in the rectangle where the luminance or the luminance variance is minimum. The tire chain detection device according to claim 3, wherein a rectangle at a position is extracted as the region image. 5. The tire chain detection device according to claim 1, wherein the determination region extracting unit is configured to extract a rectangle at a predetermined position of the input image as the region image. 6. A studless tire detection device for detecting a studless tire mounting state of a vehicle and presenting information of the detection result, wherein an image inputting means for taking an image of a predetermined place as an input image and an illumination. An image storage unit that stores the input image, and detects that the tire of the vehicle has arrived at the predetermined location, generates a detection signal, and indicates that the tire is captured in the input image. A tire position detection unit, a determination region extraction unit that receives the detection signal, and extracts a region image of a predetermined size including the tire from the input image stored in the image storage unit; The region image is analyzed, and fine grooves on the surface of the tire are extracted as characteristics of the studless tire, and the studless tire mounting state of the vehicle is determined based on the characteristics. A tire pattern determination unit, wherein the tire pattern determination unit for determining an information presentation unit for performing the information presentation based on the results, studless tire detecting apparatus characterized by comprising. 7. A studless tire detection device for detecting a studless tire mounting state of a vehicle and presenting information of the detection result, wherein a photographing direction is changed based on given position information, Image input means for inputting an image captured by changing the zoom and focus based on the input image, an image input unit having illumination for changing a light irradiation direction based on the given position information, and An image storage unit for storing, a tire position detecting unit for detecting a tire position of the vehicle, transmitting the position information indicating the tire position to the image input unit, and generating a detection signal indicating that the tire position is detected. Receiving the detection signal and extracting a region image of a predetermined size including the tire from the input image stored in the image storage unit. A region extracting unit, analyzing the extracted region image, extracting fine grooves on the surface of the tire as features of the studless tire, and determining a tire studless tire mounting state of the vehicle based on the feature. A studless tire detection device, comprising: a unit; and an information presenting unit that presents the information based on a determination result of the tire pattern determining unit. 8. The determination area extraction unit detects a ground contact surface of the tire from a luminance histogram of the input image, and extracts the rectangular region image of the tire portion in the input image based on the contact surface. The studless tire detection device according to claim 6, wherein the studless tire detection device is configured. 9. The determination area extracting unit moves a rectangle having the same size as the area image along the ground plane, and detects a position in the rectangle where the luminance or the luminance variance is minimum. The tire chain detection device according to claim 8, wherein a rectangle at a position is extracted as the region image. 10. The tire chain detection device according to claim 6, wherein the determination region extracting unit is configured to extract a rectangle at a predetermined position of the input image as the region image. 11. A tire pattern determination unit, comprising: a binarization processing unit that binarizes the area image to obtain a binary image; an edge detection unit that extracts an edge in the binary image to obtain an edge image An edge density calculation unit that calculates an edge density in the tire from white pixels of the binary image and white pixels of the edge image; and a determination unit that determines whether the tire is a studless tire based on the edge density. 11. The studless tire detecting device according to claim 6, wherein the studless tire detecting device is configured as follows. 12. The tire pattern determination unit, a division processing unit that divides the area image into a plurality of parts to obtain a small area image, and selects a threshold from the luminance of each of the small area images, and uses the threshold. A binarization processing unit that binarizes each of the small area images to obtain a plurality of binary images; an edge detection unit that extracts the edges in each of the binary images to obtain a plurality of edge images; An edge density calculation unit that determines an edge density in the tire from white pixels of a plurality of entire edge images and white pixels of the region image; and a determination unit that determines whether the tire is a studless tire from the edge density. 11. The studless tire detecting device according to claim 6, wherein the studless tire detecting device is configured as follows. 13. A tire pattern determination unit, comprising: a binarization processing unit that binarizes the area image to obtain a binary image; and an edge detection unit that detects an edge in the binary image to obtain an edge image. A block separation unit configured to replace a small number of black pixels in a region composed of substantially white pixels in the binary image with white pixels to separate the image into blocks of a white pixel group; A mask processing unit that extracts a portion of the block from the region image as a mask image; an edge detection unit that detects an edge in the region image of the extracted block portion to obtain an edge detection image; Edge density calculation for calculating an edge density in the tire from white pixels of the image separated into blocks and white pixels of the edge detection image obtained by the edge detection unit The studless tire detection device according to claim 6, 7, 8, 9, or 10, wherein the studless tire detection device is configured by: a protrusion; and a determination unit that determines whether the tire is a studless tire based on the edge density. . 14. A configuration in which the binarization processing section obtains a luminance histogram of an area to be processed, detects a position of two peak values, and sets a minimum value between the peak values as the binarization threshold. 13. The method according to claim 11, wherein:
3. The studless tire detection device according to 3.