JP2009266136A - Road structure abnormality detector - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a road structure abnormality detector can perform a wide range of monitoring by an imaging means only without additional sensors, and effectively detect abnormality of road structures at low costs. <P>SOLUTION: Each detecting zone is set up on images photographed with a camera at designated intervals by a detecting zone setting section 11. A day and night determination section 12 determines the day and night by image-processing about a day and night determination zone among those zones. When a determined result is determined at night, an illumination damage detection section 13 detects damaged illumination by making image-processing of an illumination damage detection zone. When the determined result for day and night determination is at day, a pole sign damage detection section 14 detects damage on a pole and a sign by making image-processing of a pole and sign damage detection zone. Further, a guardrail damage detection section 15 detects damage on a guardrail by making image-processing of a guardrail damage detection zone. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a road structure abnormality detection device that is used while being fixed to a road, detects a breakage state of a sign, a guardrail, a lighting fixture, and the like on the road and notifies an administrator.

In the conventional road abnormal object detection, there exist some which detect an abnormal object using things other than the image | video from a camera. For example, Patent Document 1 discloses a technique for monitoring an object in a monitoring area using a television camera and a radio wave sensor.
Japanese Patent Application Laid-Open No. 2004-151561 discloses a millimeter wave radar device that detects a fallen object and a moving target separately from guardrails and road signs.

JP 2002-222487 A (pages 3 to 4, FIG. 1) JP 2000-105276 A (3rd to 4th pages, FIG. 1)

  However, if a radio wave sensor or a millimeter wave radar device is used as in Patent Literature 1 or Patent Literature 2, a wide range of monitoring can be performed at low cost because of the narrow effective range of the radio wave sensor or millimeter wave radar device. There was a problem that it was difficult.

  The present invention has been made in order to solve the above-described problems, and does not require an additional sensor, performs wide-area monitoring with only an imaging means, and is efficient in low-cost and efficient road structures. An object of the present invention is to obtain a road structure abnormality detection device capable of detecting

  In the road structure abnormality detection device according to the present invention, an image capturing unit that acquires a road state captured at a predetermined cycle, a day / night determination region for determining day and night on an image acquired by the image capturing unit, and A detection area setting means for setting an abnormality detection area for detecting an abnormality of a structure on the road, a day / night determination means for determining day and night by processing an image of a day / night determination area set by the detection area setting means, and According to the determination result of the day / night determination means, an abnormality detection area on the image is processed, and a structure abnormality detection means for detecting an abnormality of the structure on the road is provided.

  As described above, the present invention provides an image capturing unit that acquires a road state captured at a predetermined cycle, a day / night determination region for determining day and night on an image acquired by the image capturing unit, and a road. Detection area setting means for setting an abnormality detection area for detecting an abnormality of the structure, day / night determination means for processing the image of the day / night determination area set by the detection area setting means to determine day and night, and this day / night determination According to the determination result of the means, the image processing is performed on the abnormality detection area on the image, and the structure abnormality detection means for detecting the abnormality of the structure on the road is provided. By this, it becomes possible to detect simply and efficiently.

Embodiment 1 FIG.
Embodiment 1 will be described with reference to the drawings.
1 is a block diagram showing a configuration of a road structure abnormality detection device according to Embodiment 1 of the present invention.
In FIG. 1, an image capturing unit 10 (image capturing unit) acquires an image captured at a predetermined cycle by the image sensor of the ITV camera, and stores it in a storage unit in the image capturing unit 10 as digital data. At this time, if the image captured by the image sensor is analog data, it is changed to digital data and stored. The detection area setting unit 11 (detection area setting means) sets a detection area of each detection unit, which will be described later, on the image. There are four types of setting areas, a day / night determination area A for determining day and night, an illumination breakage detection area B for detecting breakage of illumination, and a pole / signature breakage detection area C for detecting breakage of poles and signs, This is a guardrail breakage detection region D for detecting breakage of the guardrail.
The day / night determination unit 12 (day / night determination means) determines day and night in the day / night determination region A on the image set by the detection region setting unit 11.

When the illumination breakage detection unit 13 (illumination breakage detection means) is determined to be night by the day / night determination unit 12, is the illumination in the illumination breakage detection region B set by the detection region setting unit 11 turned on at night? Detect if.
The pole / signature breakage detection unit 14 (pole sign breakage detection means), when the day / night judgment unit 12 determines that the day is noon, about the image in the pole / signature breakage detection region C set by the detection region setting unit 11 Determine if poles or signs are damaged.
The guardrail breakage detection unit 15 (guardrail breakage detection means) detects the guardrail breakage from the image set in the guardrail breakage detection region D by the detection region setting unit 11 when the day / night determination unit 12 determines that it is noon.
The illumination breakage detection unit 13, the pole / marker breakage detection unit 14, and the guardrail breakage detection unit 15 are structural abnormality detection means.

FIG. 2 is a road condition diagram acquired by the image capturing means of the road structure abnormality detection device according to Embodiment 1 of the present invention.
In FIG. 2, a guardrail 2, a sign 3 attached to the upper part of the pole 5, and an illumination 4 attached to the upper part of the pole 6 are arranged on the road having the center line 1.

FIG. 3 is a situation diagram showing a region setting situation of the detection region setting unit of the road structure abnormality detection device according to Embodiment 1 of the present invention.
In FIG. 3, 1 to 6 are the same as those in FIG. The detection area setting unit 11 has a rectangular day / night determination area A at a position where there is no road structure in the upper left corner of the screen, a rectangular illumination breakage detection area B at a position where illumination is present, and a rectangular pole The sign breakage detection region C and a guardrail breakage detection region D having an arbitrary shape are set so as to cover the guardrail at a position where the guardrail exists.

Next, the operation will be described.
In the first embodiment, the presence or absence of breakage of a lighting fixture, a pole, a sign, or a guard rail is simply and efficiently detected only by image processing.
The image capturing unit 10 acquires an image showing a road situation as shown in FIG. 2 captured by a camera at a predetermined cycle and stores it as digital data.
The detection area setting unit 11 sets an area to be detected on the image. Specifically, as shown in FIG. 3, day / night determination area A for determining day and night, illumination breakage detection area B set in the illumination portion to detect breakage of illumination, breakage of poles and signs A pole / marker breakage detection area C set in a part with a pole and a sign to detect the failure, and a guardrail breakage detection area D set in a part with a guardrail to detect breakage of the guardrail are set.

When the area to be detected is set on the image, the day / night determination unit 12 monitors the average luminance value of the pixels in the day / night determination area A on the image. If it becomes, it will be judged as noon.
In the following, each detection is performed based on this determination.
The illumination breakage detection unit 13 operates when the day / night determination unit 12 determines that the night is night, and converts whether the illumination in the illumination breakage detection region B is lit at night by the average luminance value in the illumination breakage detection region B. If it is equal to or greater than a certain threshold, it is determined that the illumination is on, and if it is equal to or less than the certain threshold, it is determined that the illumination is off or cannot be lit.

  The pole / marker breakage detection unit 14 operates when the day / night determination unit 12 determines that it is noon, and performs a Hough transform on the image in the pole / marker breakage detection region C set by the detection region setting unit 11. If the straight line shape is sufficiently long with respect to the area frame, it is judged that there is no breakage and is healthy. Is determined to be damaged.

Next, the guardrail breakage detection unit 15 operates when the day / night determination unit 12 determines that it is noon. A normal image in the guardrail breakage detection area D is captured in advance as a texture, and is used as a reference template image. The correlation between the image textures is normalized using the reference template image and the image in the same coordinate area of the input image, and the edges of these two images are extracted to compare the correlation between the edge distributions. If one correlation value is high, it is determined that the guardrail is not damaged, and if the correlation value is low, it is determined that some damage is observed in the guardrail.
Here, the reason why the guardrail breakage detection unit 15 does not use the same method as the pole / marking breakage detection unit 14 is that the pole and the sign basically include a lot of straight lines, and if it breaks, it bends and deforms. Although it is possible to detect the guardrail, it is assumed that the guardrail basically includes a curve. Therefore, a method different from that of the pole / marker breakage detection unit 14 is used.

  According to the first embodiment, it is possible to easily and efficiently detect the presence or absence of damage to road structures such as lighting fixtures, poles, signs, guardrails, and the like only by image processing.

Embodiment 2. FIG.
FIG. 4 is a block diagram showing a configuration of a road structure abnormality detection apparatus according to Embodiment 2 of the present invention.
In FIG. 4, 10 to 15 are the same as those in FIG. In FIG. 4, in front of the day / night determination unit 12, the traffic angle determination unit 16 that newly determines the traffic jam on the road, and the camera angle of view change based on the instruction to change the camera field angle from the traffic congestion determination unit 16, A camera control unit 17 that notifies the detection area setting unit 11 of a change in the camera angle of view is provided. The detection area setting unit 11 changes the detection area so that it matches the angle of view of the camera whose angle of view has been changed by the camera control unit 17. Here, the angle of view refers to the direction of the camera and the zoom magnification.
In the second embodiment, a preset registration type zoom turning camera is used as the camera. This preset registration type zoom turning camera is a camera in which a plurality of angles of view, that is, camera directions and zoom magnifications can be registered in advance, and can be sequentially changed to the plurality of registered angles of view. In the second embodiment, the registered angle of view (each having a preset number) is changed by instructing the preset number from the camera control unit 17.

Next, the traffic jam determination unit 16 and the camera control unit 17 will be described with reference to FIG.
Usually, in road monitoring using image processing, the traffic situation on the road is mostly monitored. For this reason, the road structure abnormality detection device for detecting an abnormality of the road structure according to the present invention is provided with the function of a traffic state monitoring device in the second embodiment.

  The traffic jam determination unit 16 (traffic jam determination means) determines the presence or absence of traffic jams using the vehicle occupancy with respect to the entire road surface on the road lane and the vehicle average speed. Here, the period of whether or not there is a traffic jam is further counted, and the camera control unit 17 is instructed to change the angle of view of the camera in a quiet period where the non-traffic period lasts for a certain period.

  In addition to the angle of view that is basically set, the camera can monitor other angles of view, such as a reverse lane, by using a preset registration type zoom turning camera. In addition to this, it is possible to detect a more detailed breakage state by zooming the road structure in the region where the camera can turn. For this reason, the camera control unit 17 changes the angle of view of the camera, transmits the changed camera angle of view information to the detection area setting unit 11, and the detection area setting unit 11 performs detection according to the angle of view of the camera. Set the area.

The traffic judgment unit 16 is used in normal operation and allocates most of the functions during operation to traffic state monitoring. When detecting an abnormality in the structure, it is not necessary to always operate the traffic jam determination unit 16, and verification may be performed discretely. Therefore, when the non-congested period continues for T hours or longer, the camera control unit 17 is urged to change the preset angle of view.
When changing the preset angle of view of the camera, the camera control unit 17 changes a parameter for setting a detection area for road abnormality detection for the angle of view. Therefore, preset angle of view change information is output to the detection area setting unit 11.
The camera control unit 17 sequentially sets all of the registered preset view angles at a predetermined timing, performs abnormality detection by this, and returns to the normal basic setting view angle when the abnormality detection process ends. Return to congestion monitoring.

  According to the second embodiment, it is usually determined whether there is traffic in the monitoring road lane, which is the basic processing of image processing monitoring, and abnormality detection of road structures is detected in the low season of the vehicle. By using the camera to zoom the object of the road structure, it is possible to determine whether there is damage in more detail.

Embodiment 3 FIG.
FIG. 5 is a block diagram showing a configuration of a road structure abnormality detection device according to Embodiment 3 of the present invention.
In FIG. 5, 10 to 17 are the same as those in FIG. In FIG. 5, a camera switching unit 18 (camera switching unit) that switches videos of a plurality of cameras in response to a camera switching signal from the camera control unit 17 is provided.

FIG. 6 is a block diagram showing connection of a camera switching unit using an analog image of the road structure abnormality detection device according to Embodiment 3 of the present invention.
In FIG. 6, n systems of input analog video from n cameras are input to the camera switching section 18, and one system of input analog video switched by the camera switching signal from the camera control section 17 is input to the image capturing section 10. Output. The camera control unit 17 controls camera switching by the camera switching unit 18.

FIG. 7 is a block diagram showing connection of a camera switching unit using a digitally transmitted image of the road structure abnormality detection device according to Embodiment 3 of the present invention.
In FIG. 7, the camera switching unit 18 handles digital data input from the camera through a network line and provides two moving image data expansion devices. The moving image data expansion device 21 for the camera currently being processed and the next cycle processing are provided. The moving image data expansion device 22 for the camera is provided. Switching between the moving image data expansion device 21 for the camera currently being processed and the moving image data expansion device 22 for the camera in the next cycle processing is performed by a camera switching signal from the camera control unit 17. The digital data transmitted through the network line includes the data for the moving image data expansion device 21 for the camera currently being processed and the moving image data expansion device 22 for the camera of the next cycle processing.

Next, the operation of the third embodiment will be described.
In the second embodiment, damage detection is performed using a preset registration type zoom turning camera. However, in the third embodiment, damage detection is performed using a plurality of cameras based on the control of the camera control unit 17. Is to do.
When handling camera images, one analog signal is assigned to one camera. Therefore, in order to monitor a plurality of camera images, as many image capturing units as the number of cameras are required. Here, the image capturing unit is a part that performs analog-to-digital conversion and digitizes the video. However, when a plurality of such devices are used, the device scale increases.
For this reason, in the third embodiment, the images of a plurality of cameras are temporarily aggregated by the camera switching unit 18, and the camera switching unit 18 performs switching so as to take out one necessary camera video. To supply video. The camera control unit 17 instructs the camera switching unit 18 to switch. For example, the camera switching unit 18 instructs the camera switching at regular intervals or manually.
Thereby, by setting the camera to be processed by the camera switching unit 18, it is possible to change the setting of the detection area of the subsequent traffic jam determination unit 16 or the detection area setting unit 11 as necessary.

In recent wide-area video surveillance systems, surveillance video may be handled as digital data instead of analog. Therefore, as shown in FIG. 7, the camera switching unit 18 handles digital data.
In the wide-area video surveillance system, video is transmitted in the network by a compressed moving image called MPEG2. For this reason, in image processing that handles continuous images of still images, it is necessary to develop compressed moving image data.
Since the decompression work of the compressed data requires a certain period, there is a non-monitoring period in the video switching period. Therefore, the compressed moving image data expansion device is duplicated, and the video expansion work of the camera to be switched next is started in advance, thereby eliminating the non-monitoring period and efficiently switching the camera video.
The digital data transmitted through the network line includes a mixture of the moving image data expansion device 21 for the camera currently being processed and the moving image data expansion device 22 for the camera of the next period processing.

  According to the third embodiment, it is possible to provide a road structure abnormality detection device to an administrator at a lower cost by monitoring a plurality of camera images with a single device (image capturing unit). It becomes.

It is a block diagram which shows the structure of the road structure abnormality detection apparatus by Embodiment 1 of this invention. It is the road condition figure acquired by the image pick-up means of the road structure abnormality detection apparatus by Embodiment 1 of this invention. It is a situation figure showing the area setting situation of the detection area setting part of the road structure abnormality detection device by Embodiment 1 of this invention. It is a block diagram which shows the structure of the road structure abnormality detection apparatus by Embodiment 2 of this invention. It is a block diagram which shows the structure of the road structure abnormality detection apparatus by Embodiment 3 of this invention. It is a block diagram which shows the connection of the camera switching part using the analog image of the road structure abnormality detection apparatus by Embodiment 3 of this invention. It is a block diagram which shows the connection of the camera switching part using the digital transmission image of the road structure abnormality detection apparatus by Embodiment 3 of this invention.

Explanation of symbols

1 road center line, 2 guardrails, 3 signs, 4 lighting, 5 and 6 poles,
DESCRIPTION OF SYMBOLS 10 Image pick-up part, 11 Detection area setting part, 12 Day / night determination part, 13 Illumination breakage detection part,
14 pole / signature breakage detection part, 15 guardrail breakage detection part, 16 traffic jam judgment part,
17 camera control unit, 18 camera switching unit, 21 moving image data expansion device 1, 22 moving image data expansion device 2.

Claims (7)

  Image capturing means for acquiring a road state imaged at a predetermined cycle, day / night determination area for determining day and night on the image acquired by the image capturing means, and abnormality of a structure on the road According to the detection area setting means for setting the abnormality detection area, the day / night determination means for processing the image of the day / night determination area set by the detection area setting means to determine day and night, and the determination result of the day / night determination means, A road structure abnormality detection device comprising a structure abnormality detection means for performing image processing on an abnormality detection region on the image and detecting an abnormality of the structure on the road.   2. The road structure abnormality detection device according to claim 1, wherein the structure abnormality detection means includes illumination breakage detection means for detecting breakage of lighting equipment when the determination by the day / night determination means is night.   3. The road structure abnormality detecting device according to claim 1, wherein the structure abnormality detecting means includes guard rail breakage detecting means for detecting breakage of the guardrail when the determination by the day / night determining means is daytime. Detection device.   The said structure abnormality detection means includes a pole sign breakage detection means for detecting breakage of the pole and the sign when the determination by the day and night determination means is daytime. The road structure abnormality detection device described in 1.   The traffic flow measurement according to any one of claims 1 to 4, further comprising: a traffic jam judging means for judging the presence or absence of traffic jam from a road surface occupation ratio and a vehicle traffic average speed of a vehicle passing through a road lane. apparatus.   The road structure abnormality detection device according to any one of claims 1 to 5, wherein the image acquired by the image pickup means is picked up using a preset registration type zoom turning camera.   The image acquired by the image capturing unit is captured by a plurality of cameras, and further includes a camera switching unit that switches an output of the plurality of cameras. The road structure abnormality detection device according to claim 1.

Images