JP2003323689A - Special vehicle detecting device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a special vehicle detecting device and method and system for detecting a specific vehicle. <P>SOLUTION: A specific vehicle detecting device 20 for detecting a specific vehicle 4 on a road 2 is provided with an image processing part 23 having an extracting function for extracting special display indicating the presence of a special vehicle 4 from a photographic picture obtained by photographing the road 2 and a detecting function for detecting the special vehicle 4 based on the special display extracted by the extracting function. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a special vehicle detection device for detecting a special vehicle, a special vehicle detection method, and a special vehicle detection system.

[0002]

2. Description of the Related Art Some roads are equipped with surveillance cameras for the purpose of monitoring the road conditions. The video image from the surveillance camera is transmitted to the control room that manages the road, and is viewed by the surveillance staff in the control room. In road management, the management of special vehicles such as vehicles carrying dangerous goods is especially important, and it is necessary for the observer to monitor the special vehicles carefully.

[0003]

However, in the conventional monitoring of the special vehicle, since the monitor is performed by the monitor visually observing the monitor, the burden on the monitor becomes great.

The present invention has been made to solve the above problems, and an object of the present invention is to provide a special vehicle detection device, a special vehicle detection method, and a special vehicle detection system which can reduce the burden on the observer. To do.

[0005]

In order to achieve such an object, a special vehicle detection device according to the present invention is a special vehicle detection device for detecting a special vehicle on a road, and an image obtained by photographing the road. It is configured to include extraction means for extracting a special display indicating that it is a special vehicle from the image, and detection means for detecting a special vehicle based on the special display extracted by the extraction means.

Further, in the special vehicle detection device according to the present invention, the extraction means groups the pixels of the photographed image based on a specific condition, detects the grouped pixel group as a label, and extracts the special display. It is characterized by doing.

Further, in the special vehicle detection device according to the present invention, the extraction means combines the labels based on a specific condition,
The feature is that the special display is extracted based on the combination result.

Further, the special vehicle detection method according to the present invention is
A special vehicle detection method for detecting a special vehicle on a road, based on an extraction step of extracting a special display indicating a special vehicle from a captured image of a road, and a special display extracted by the extraction means. , And a detection step of detecting a special vehicle.

Further, in the special vehicle detection method according to the present invention, in the extraction step, the pixels of the photographed image are grouped based on a specific condition, and the grouped pixel group is detected as a label to extract the special display. It is characterized by doing.

Further, the special vehicle detection method according to the present invention is
The extraction step is characterized in that the labels are combined based on a specific condition, and the special display is extracted based on the combination result.

The special vehicle detection system according to the present invention is a special vehicle detection system for detecting a special vehicle on a road, and a photographing means for photographing the road to obtain a photographed image,
The above-mentioned special vehicle detection device is provided.

According to these aspects of the invention, the special vehicle can be properly detected by extracting the special display which is a display such as "danger" or "poison" attached to the body of the special vehicle. Therefore, the burden on the observer can be reduced.

In addition, in the special vehicle detection system according to the present invention, the special vehicle detection system is installed before the special vehicle entry prohibited section, and based on the detection result of the special vehicle detection device, the special vehicle driver is prohibited from entering. Is further provided with an informing means for informing.

In this case, since the driver of the special vehicle can be informed that the special vehicle is prohibited from entering, it is possible to prevent the special vehicle from entering the prohibited section. .

[0015]

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

A special vehicle detection device, a special vehicle detection method, and a special vehicle detection system according to the embodiment will be described.

FIG. 1 is a block diagram of a special vehicle detection system according to this embodiment. As shown in FIG. 1, the special vehicle detection system 1 according to the present embodiment includes a special vehicle 4 on a road 2.
Is used to check the presence or absence of the special vehicle 4 in the tunnel 3 and to be useful in dealing with a disaster, for example. This special vehicle detection system 1
Includes cameras 10a to 10f, a special vehicle detection device 20, a display board 30, and a management center 40.

The cameras 10a to 10f are photographing means for photographing the road 2 to obtain photographed images. Cameras 10a, 10
b is installed on the entrance side of the tunnel 3 in order to detect the special vehicle 4 attempting to enter the tunnel 3. The cameras 10c and 10d are installed near the entrance of the tunnel 3, and the cameras 10e and 10f are installed near the exit of the tunnel 3. These cameras 10c to 10f are for grasping the special vehicle 4 in the tunnel 3. Since the special indications such as "danger" and "poison" attached to the special vehicle 4 are provided on at least one of the front part and the rear part of the special vehicle 4, these cameras 10a to 10f are each set in two sets. The special vehicle 4 can be photographed from the front and the rear. Cameras 10a-10
f is connected to the special vehicle detection device 20, and transmits the photographed image to the special vehicle detection device 20.

The special vehicle detection device 20 includes an A / D converter 2
1, an image memory 22 and an image processing unit 23. The A / D conversion unit 21 performs A / D conversion of an input captured image signal or the like from an analog signal to a digital signal, and for example, a video A / D converter or the like is used. If the cameras 10a to 10f output digital signals, the installation of the A / D converter 21 may be omitted.

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

The image processing unit 23 performs image processing on the image data stored in the image memory 22, extracts the special display provided on the special vehicle 4, and detects the special vehicle 4 based on the extracted special display. It is a thing. That is, the image processing unit 23 detects the special vehicle 4 based on the extraction function of extracting the special display indicating the special vehicle 4 from the captured image of the road 2 and the special display extracted by the extraction function. It is equipped with a detection function. In addition, the image processing unit 2
3 is connected to the display board 30 and the management center 40. The image processing unit 23 transmits the notification content based on the detection result to the display board 30 and the detection result to the management center 40.

The display board 30 is installed in front of the prohibition section of the special vehicle 4 and functions as a notification means for notifying the driver of the special vehicle 4 that the special vehicle 4 is prohibited. For example,
The display board 30 receives the notification content based on the detection result detected by the special vehicle detection device 20, displays the content, and notifies that entry is prohibited.

The management center 40 has a monitoring monitor 41 and an operation support monitor 42 in the control room. The monitoring monitor 41 displays captured images captured by the cameras 10a to 10f. In addition, the operation support monitor 42
It has a function of displaying and outputting (for example, printing out) the detection result of the special vehicle detection device 20. Further, the operation support monitor 42 is connected to the cameras 10a to 10f and has a function of adjusting parameters regarding these detection accuracy.

Next, the operations of the special vehicle detection system 1 and the special vehicle detection device 20 will be described.

First, the road 2 is taken by the cameras 10a to 10f.
Will be taken. Then, the photographed images photographed by the cameras 10a to 10f are input to the special vehicle detection device 20 as image signals. The image signal input to the special vehicle detection device 20 is A / D converted by the A / D converter 21 from an analog signal to a digital signal. The digitally converted image signal is stored in the image memory 22 as image data. Then, the following processing is performed by the image processing unit 23 according to the flowchart shown in FIG.

First, in S1 of FIG. 2, the image memory 2
The image data stored in 2 is read, and binarization processing is performed based on the brightness value of each pixel.

Then, the flow shifts to S2, where the pixels of the photographed image are grouped on the basis of a specific condition, and the label extraction processing is performed with the grouped pixel group as a label. Here, adjacent pixels are grouped based on the result of the binarization process, and the grouped pixel group is used as a label. The situation is shown in FIG. In addition, in FIG.
One side of the square is 10 pixels, and the number of pixels in one square is 100.
I am trying. Further, in the following description with reference to FIG.
A cell is treated like one pixel.

As shown in the figure, the pixels P ₁ and P ₂ are grouped because they belong to the high-luminance side and are adjacent to each other as a result of the binarization process. Similarly, other pixels belonging to the high brightness side are also grouped to form the label L ₀ . The labels L _{1 to} L ₆ are also grouped in the same manner as the label L ₀ .

At this time, the width of each label L _{0 to} L ₆ is
Height and area are calculated. Here, when the horizontal direction of the captured image is the X direction and the vertical direction of the captured image is the Y direction, the width is the pixel in the X direction in the circumscribing frame of each label L _{0 to} L _6. The number is the number, and the height is the number of pixels in the Y direction in the circumscribing frame of each label L _{0 to} L ₆ , and the area is the number of pixels in the circumscribing frame of each label L _{0 to} L ₆ .
Thus, for label L ₀ , the width is 60 and the height is 9
0 and the area is 1700. Here, the circumscribing frame is a frame that circumscribes the labels L _{0 to} L ₆ in the minimum area, and has, for example, a rectangular shape.

After that, when the label extracting process in S2 of FIG. 2 is completed, the process proceeds to S3, and the label combining process for combining the labels L _{0 to} L _{6 based} on a specific condition is performed. This state is shown in FIG. It should be noted that FIG. 4 is similar to FIG. 3, and in FIG. 3, one side of one square is 10 pixels, and the number of pixels in one square is 100. Further, in the following description with reference to FIG. 4, one square is treated like one pixel.

In FIG. 4, label L ₀ and label L ₁ , and label L ₂ and label L ₃ are combined.
For example, the combined label L ₀₁ in which the label L ₀ and the label L ₁ are combined is formed in the character “danger”. By performing the combining process in this manner, it is possible to extract an appropriate special display.

Next, in S4, candidates are created. The creation of this candidate can be matched with the combined label L ₀₁ , L ₂₃ and the label L ₄ , L ₅ which cannot be matched with the special display of the special vehicle 4 based on a specific condition. It is a process of separating things from each other. By this process,
Those that may be special displays are extracted.

After that, the processing shifts to S5, and character recognition processing is performed. Here, in S4, the characteristic amount peculiar to the separated candidates (each character) is obtained, the evaluation amount is calculated by the discriminant function with the dictionary data of each character that is obtained in advance, and the character is determined by the evaluation amount. That is, the processes of S2 to S5 correspond to the extraction process of extracting the special display indicating the special vehicle 4 from the captured image of the road 2. Then, the special vehicle 4 is detected based on the extracted special display.

When the special vehicle 4 is detected on the basis of the picked-up images picked up by the cameras 10a and 10b, the image processing section 23 sends the information of the detection result to the display plate 30, and the picked-up image and the detection result are detected. Information is management center 4
Sent to 0. On the display board 30, it is displayed that the entry is prohibited based on the information of the transmitted detection result.
As a result, the driver of the special vehicle 4 can be notified that entry is prohibited. In the management center 40, the captured image is displayed on the monitoring monitor 41, and the information on the detection result is displayed on the operation support monitor 42. The observer can easily confirm from the information displayed on the operation support monitor 42 that the special vehicle 4 is about to enter the prohibited section.

When the special vehicle 4 is detected on the basis of the images taken by the cameras 10c to 10f,
The image processing unit 23 transmits the captured image and the detection result information to the management center 40. In the management center 40,
The captured image is displayed on the monitoring monitor 41, and the information on the detection result is displayed on the operation support monitor 42. The observer can easily know the number of special vehicles 4 in the tunnel 3 from the information displayed on the operation support monitor 42. In this way, the entry of the special vehicle 4 into the prohibited area, the number of vehicles in the tunnel 3 and the like can be easily checked, so that the burden on the observer is reduced.

Next, the label combining process (S3) of FIG. 2 will be described in detail with reference to FIGS.

First, in S31 of FIG. 5, a deletion process is performed. This is a process for deleting an impossible special display, and is a process for deleting the label L ₆ shown in FIG. This process is performed by, for example, each label L _{0 to} L ₆
Based on the area of. That is, each label L ₀
From the areas of L ₆ to L ₆ , those less than the area 350 are extracted and deleted.

Then, the flow shifts to S32 in FIG. 5, in the X direction.
About another label L within the specified range ₀~ L₆Exists
A determination is made as to whether or not. That is, the label L₀Say about
For example, label L in the X direction₀Within the specified range from others
Label L₁~ L₆Is determined.
Here, another label L₁~ L₆Whether or not exists
Is performed based on the number of pixels in the X direction, for example. sand
Wow, each label L₀~ L₆X direction (+ side,-
On both sides), the other label L within 5 pixels₀~ L₆Exist
It is determined whether or not it exists. Specifically, each label L₀
~ L₆Other labels within 5 pixels in the X direction from the circumscribed frame of
L₀~ L₆It is determined whether or not the circumscribing frame is present. here
And another label L₀~ L₆If is present, go to S33.
Then, the label combining process is performed. Here, all la
Bell L₀~ L₆Of two adjacent pixels within 5 pixels
Be combined. After the combination, the process proceeds to S34.

On the other hand, if it is determined that the other labels L _{0 to} L ₆ do not exist within the predetermined range in the X direction, the process proceeds to S34 without performing S33. Note that, in FIGS. 3 and 4, since there are no other labels L _{0 to} L ₆ within the number of pixels 5 in the X direction, in this case, the label combining process is not performed.

In S34 of FIG. 5, it is judged whether or not the other labels L _{0 to} L ₆ are present within a predetermined range in the Y direction. That is, when it comes to the label L _0, Y direction other labels L ₁ ~ from the label L ₀ within a predetermined range for
A determination is made as to whether L ₆ is present. Here, the determination as to whether or not the other labels L _{1 to} L ₆ are present is performed by, for example, Y
It is performed based on the number of pixels in the direction. That is, in the Y direction (both the positive side and the negative side) of each of the labels L _{0 to} L ₆ ,
It is determined whether or not there are other labels L _{0 to} L ₆ within the number of pixels of 15. Specifically, whether the circumscribing frame of each label L ₀ ~L from bounding box of another within the pixel number 15 in the Y direction of ₆ labels L ₀ ~L ₆ is present or not. Here, if there are other labels L _{0 to} L ₆ , the process proceeds to S35 and the label combining process is performed. Here, all labels L _{0 to} L ₆
Among them, two adjacent ones within 15 pixels are combined. After combining, the label combining process ends.

On the other hand, when it is determined that the other labels L _{0 to} L ₆ do not exist within the predetermined range in the Y direction, the label combining process ends without performing the process of S35. In addition,
3 and 4, regarding the label L ₀ , another label L ₁ exists within the number of pixels 15 in the Y direction, and regarding the label L ₂ , the other label within the number of pixels 15 within the Y direction. Since L ₃ exists, both of them (label L ₀ , label L ₁ , label L
₂ and the label L ₃ ) are combined.

Next, the candidate creation process (S4) of FIG. 2 will be described in detail with reference to FIGS.

First, in S41 of FIG. 6, it is determined whether the area is larger than the lower limit value. The binding labels L ₀₁ and L ₂₃ are the targets for determining whether or not this is larger.
And labels L _{0 to} L ₅ . The lower limit value here is, for example, the area 2000, and when it is determined that the area is not larger than 2000, the process proceeds to S44, and the label or the combined label is deleted. On the other hand, the area is 2000
If it is determined that it is larger, the process proceeds to S42.

As shown in FIG. 4, the combined label L ₀₁
Has an area of 3100, and the combined label L _{23 has} an area of 16
00, and the areas of the labels L ₄ and L ₅ are 400. Therefore, in the processing of S41 of FIG. 6, only the combined label L ₀₁ is
It is determined that the area is larger than 2000, and the process proceeds to S42,
Others are determined not to be larger than the lower limit value, the process proceeds to S44 and is deleted.

In S42, it is determined whether the area is smaller than the upper limit value. This process is also determined in the same manner as the process of S41. The upper limit value here is, for example, 10000, and when it is determined that the area is not smaller than 10000, the process proceeds to S44 and the label or the combined label is deleted. On the other hand, if it is determined that the area is smaller than 10,000, then the flow shifts to S43.

In S43, it is determined whether or not the ratio of height to width is smaller than a predetermined value. If the predetermined value here is 35, for example, the determination here is made based on the following equation.

[0047] abs (100- (height ÷ width × 100)) (1)

If the value obtained by the equation (1) is not smaller than 35, the process proceeds to S44 and is deleted.

On the other hand, when the value obtained by the equation (1) is smaller than 35, the process shifts to S45, and the label is registered as a candidate as a special display. And
The candidate creation process ends.

In the explanation of FIG. 6, the combined labels L ₀₁ , L ₂₃ and the labels L _{0 to} L ₅ are judged one by one and S is determined.
In the preceding stage of the processing of 45, S41 to S43 for all labels
The process may be looped by determining whether or not the process has been executed.

As described above, according to the present embodiment, the special vehicle 4 can be properly controlled by extracting the special indications such as “danger” and “poison” attached to the vehicle body of the special vehicle 4. Can be detected. Therefore, the burden on the observer can be reduced.

Further, for example, it is possible to detect the special vehicle 4 based on the vehicle number, but according to this embodiment, the special vehicle is displayed based on the special display which is displayed larger than each character of the vehicle number. Since 4 is detected, the detection accuracy can be higher than in the case of detecting the special vehicle 4 based on the vehicle number.

Further, according to this embodiment, the driver of the special vehicle 4 can be notified that entry is prohibited. Furthermore, for example, the prohibited area is the special vehicle 4
If it is difficult or dangerous for the passenger to pass, an accident can be prevented.

Further, according to this embodiment, the driver of the special vehicle 4 can be notified that the special vehicle 4 is prohibited from entering, so that the special vehicle 4 is prevented from entering the prohibited section. Can be prevented.

Further, according to the present embodiment, the existence of the special vehicle 4 in the no-entry section or the tunnel can be recognized.

[0056]

As described above in detail, according to the present invention, since a special vehicle can be detected, it is possible to reduce the burden on the surveillance staff.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a special vehicle detection system according to the present embodiment.

FIG. 2 is a flowchart showing a process of a special vehicle detection device 20 shown in FIG.

FIG. 3 is a diagram showing labels L _{0 to} L ₆ extracted in a label extraction process (S2) shown in FIG.

4 is a diagram showing combined labels L ₀₁ and L ₂₃ combined in the label combining process (S3) shown in FIG.

FIG. 5 is a detailed flowchart of the label binding process (S3) shown in FIG.

FIG. 6 is a detailed flowchart of a candidate creation process (S4) shown in FIG.

[Explanation of symbols]

1 ... Special vehicle detection system, 4 ... Special vehicle, 10a-1
0f ... Camera, 20 ... Special vehicle detection device, 23 ... Image processing unit, 30 ... Display board.

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04N 7/18 H04N 7/18 K F term (reference) 5B057 AA16 AA19 BA02 CE12 DA11 DA15 DB02 DB09 DC03 DC04 DC14 DC33 DC36 5B064 AA02 AA07 AB02 DB06 5C054 AA01 FC12 FE28 HA26 HA30 5H180 AA13 CC04 EE07

Claims (8)

[Claims] 1. A special vehicle detection device for detecting a special vehicle on a road, comprising: extracting means for extracting a special display indicating the special vehicle from a captured image of the road; and the extracting means. A special vehicle detection device comprising: a detection unit that detects the special vehicle based on the extracted special display. 2. The extracting means groups the pixels of the photographed image based on a specific condition, detects the grouped pixel group as a label, and extracts the special display. Item 1. The special vehicle detection device according to item 1. 3. The special vehicle detection device according to claim 2, wherein the extraction unit combines the labels based on a specific condition and extracts the special display based on a result of the combination. 4. A special vehicle detection method for detecting a special vehicle on a road, comprising: an extracting step of extracting a special display indicating the special vehicle from a captured image of the road; and the extracting means. A detection step of detecting the special vehicle based on the extracted special display, and a special vehicle detection method. 5. The extracting step is characterized in that the pixels of the photographed image are grouped based on a specific condition, the grouped pixel group is detected as a label, and the special display is extracted. Item 4. The special vehicle detection method according to item 4. 6. The special vehicle detection method according to claim 5, wherein in the extracting step, the labels are combined based on a specific condition, and the special display is extracted based on a result of the combination. 7. A special vehicle detection system for detecting a special vehicle on a road, comprising: photographing means for photographing the road to obtain a photographed image; and the special vehicle detection according to any one of claims 1 to 3. A special vehicle detection system comprising: a device. 8. An informing unit is installed in front of the no-entry section of the special vehicle, and further provides an informing unit for informing a driver of the special vehicle that no entry is permitted based on a detection result of the special vehicle detection device. The special vehicle detection system according to claim 7, further comprising: