JP2001273587A - Road monitor system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a road monitor system by which the occurrence of a sudden event is detected with high precision without requiring a high image processing ability. SOLUTION: A controller S controls the respective parts of the system in addition to monitor cameras 1-N in accordance with the request of an operator. A monitor M displays the operation state of the system and reports images photographed by the cameras 1-N and the occurrence of the sudden event to the operator. An image processor V inputs a video signal photographed by the monitor cameras 1-N, performs an image processing to the video signal and detects the sudden event which occurs on a road to be monitored from the processing result. The sudden event which occurs on the road owing to the non-uniformity of the following amounts, speeds and the rates is detected based on the traffic amount of respective lanes, an average speed and a lane change rate, which do not required the high image processing ability when the sudden event is detected, based on the traffic amount of the respective lanes, the average speed and the lane change rate of each monitor section area.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a road surveillance system for detecting the occurrence of a sudden event such as a traffic jam or an accident using a surveillance camera.

[0002]

2. Description of the Related Art As is well known, a conventional road monitoring system photographs a road to be monitored by a monitoring camera, analyzes the photographed image by an image processing device, and generates a sudden image generated in a photographing area. Event was to be detected.

In this type of road monitoring system, installing a large number of monitoring cameras so as to capture the entire area of a road to be monitored has various problems including costs, and therefore, images are taken. There is a monitored section area and an unmonitored section area where no image is captured.

[0004] The unmonitored section area includes not only an area that is not actually photographed but also an area that is photographed but cannot detect a passing vehicle by the image analysis capability of the image processing apparatus.

[0005] For this reason, in the unmonitored section area, sudden event detection by image analysis cannot be performed. As another method, a method of estimating the state of an unmonitored section area by identifying a passing vehicle for each monitored section area and associating the identified passing vehicle with the monitored section area is also conceivable.

However, according to this method, as the number of traffic vehicles identified increases, the process of associating becomes more difficult, and when an unexpected event such as traffic congestion occurs,
Because we could n’t accurately identify the passing vehicles,
Not only high image processing capability is required, but also the reliability of detection accuracy is not high.

[0007]

In a conventional road surveillance system, an unexpected event is detected in an unmonitored section area.
Not only high image processing capability is required, but also the reliability of detection accuracy is not high.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and has as its object to provide a road monitoring system capable of detecting occurrence of an unexpected event with high accuracy without requiring high image processing capability. .

[0009]

In order to achieve the above object, according to the first aspect of the present invention, a plurality of surveillance cameras for photographing a road to be monitored are distributed and arranged. In a road surveillance system that detects an unexpected event that has occurred on a road by performing image analysis of a photographed image, a traffic passing through each monitoring section photographed by each surveillance camera based on the image photographed by each surveillance camera. Vehicle number counting means for calculating the number of vehicles, and a deviation of the number of passing vehicles between different monitoring sections is determined based on the number of passing vehicles for each monitoring section determined by the vehicle number counting means. A sudden event detecting means for detecting a section in which the event has occurred is provided.

In the road surveillance system having the above structure, the number of vehicles passing through each monitoring section photographed by each surveillance camera is determined based on the image photographed by each surveillance camera, and then the traffic between different monitoring sections is determined. Find the deviation of the number of vehicles,
The section where the sudden event has occurred is detected based on this deviation.

Therefore, according to the road surveillance system having the above-described configuration, the number of passing vehicles in each monitoring section which does not require a high image processing capability is measured, and the deviation of the number of passing vehicles between different monitoring sections is calculated. An event occurrence section can be specified. That is, even if an unexpected event occurs in an unmonitored section where no monitoring camera is arranged, the occurrence of the above-mentioned unexpected event can be detected from the deviation of the number of passing vehicles in the preceding and following monitoring sections. It is possible to detect a section where an unexpected event has occurred including an unmonitored section area without requiring any capability.

In order to achieve the above object, according to the present invention, a plurality of surveillance cameras for photographing a road to be monitored are arranged in a distributed manner, and an image photographed by the surveillance camera is displayed. In a road surveillance system that detects an unexpected event that has occurred on a road by performing image analysis, an average speed of a vehicle passing through each monitoring section photographed by each surveillance camera is determined based on an image photographed by each surveillance camera. The average speed measuring means to be obtained and the deviation of the average speed between different monitoring sections based on the average speed for each monitoring section obtained by the average speed measuring means are determined, and the section where the sudden event has occurred is detected based on the deviation. And a sudden event detecting means.

In the road surveillance system having the above configuration, the average speed of the vehicle passing through each monitoring section photographed by each surveillance camera is determined based on the image photographed by each surveillance camera. The deviation of the average speed is obtained, and the section where the sudden event has occurred is detected based on the deviation.

Therefore, according to the road surveillance system having the above-described configuration, by measuring the average speed of each monitoring section which does not require a high image processing capability and calculating the deviation of the average speed between different monitoring sections, an unexpected event can be obtained. The occurrence section can be specified.
That is, even if a sudden event occurs in an unmonitored section where no monitoring camera is arranged, the occurrence of the above-mentioned sudden event can be detected from the deviation of the average speed of the preceding and following monitored sections, so that a high image processing capability It is possible to detect a section in which an unexpected event has occurred, including an unmonitored section area, without the need for.

Further, in order to achieve the above object, according to the present invention, a plurality of surveillance cameras for photographing a road to be monitored are arranged in a distributed manner, and an image photographed by the surveillance camera is displayed. In a road surveillance system that detects an unexpected event that has occurred on a road by performing image analysis, a vehicle that changes lanes for each monitoring section photographed by each surveillance camera based on an image photographed by each surveillance camera. A lane change rate measuring means for determining a ratio for each lane; and a deviation of a lane change rate between different monitoring sections based on the lane change rate determined by the lane change rate measurement means. A sudden event detecting means for detecting a section in which the event has occurred is provided.

In the road surveillance system having the above-described configuration, the ratio of vehicles changing lanes for each monitoring section photographed by each surveillance camera is determined for each lane based on the video photographed by each surveillance camera. , The deviation of the lane change rate between different monitoring sections is determined, and the section in which the sudden event has occurred is detected based on this deviation.

Therefore, according to the road monitoring system having the above-described configuration, the calculation of the lane change rate for each monitoring section that does not require a high level of image processing capability or the deviation of the lane change rate between different monitoring sections allows calculation of an unexpected event. The occurrence section can be specified. That is, even if an unexpected event occurs in an unmonitored section where no monitoring camera is arranged, the occurrence of the above-mentioned unexpected event can be detected from the deviation of the lane change rate in the preceding and following monitored sections. It is possible to detect a section where an unexpected event has occurred including an unmonitored section area without requiring any capability.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a configuration of a road monitoring system according to an embodiment of the present invention.

Surveillance cameras 1 to N (N> 1) are CCDs
(Charge-Coupled Device) is an electronic camera that employs an imaging device, and is disposed on a highway or a general road to be monitored, and photographs a vehicle traveling on the road.

Since not all areas of the above road are photographed by the monitoring cameras 1 to N, FIG.
As shown in (1), there are a monitored section area that can be photographed at a level required for the processing capability of the image processing apparatus V described later, and other unmonitored section areas.

The image processing apparatus V receives video signals photographed by the monitoring cameras 1 to N, performs image processing on the video signals, and, based on the processing result, an unexpected event that occurs on a road to be monitored. Is to be detected.

The control device S is a man-machine interface for controlling each part of the system. Monitoring monitor M
Is a monitor for displaying an operation state of the system, and also for notifying an operator of a video captured by the monitoring cameras 1 to N or occurrence of a sudden event.

Next, a description will be given of a process of detecting a sudden event of the road monitoring system having the above configuration. FIG. 2 is a flowchart for explaining the above processing, which is mainly performed by the image processing apparatus V. The processing of steps 2a to 2c is processing performed in parallel for each of the video signals from the monitoring cameras 1 to N.

First, in step 2a, the monitoring cameras 1 to
Detecting each passing vehicle from the video signal from N,
Each detected passing vehicle is tracked, its position is measured, and the process proceeds to step 2b.

In step 2b, based on the measurement result in step 2a, the number of passing vehicles is counted for each lane of the road to be monitored, and based on the change in the position of the passing vehicle being tracked and the time required, the passing vehicle being tracked is counted. Is calculated individually. Further, based on the position change, the number of vehicles having changed lanes is counted, and the process proceeds to step 2c.

In step 2c, the average traveling speed for each lane is calculated based on the number of traffic per lane counted in step 2b and the traveling speed calculated in step 2b. The lane change rate is calculated for each lane on the basis of the number of passing lanes for each lane and the number of lane changes counted in step 2b, and step 2d
Move to

In step 2d, a traffic flow (free flow, congestion, congestion) is determined based on the number of traffic per lane counted in step 2b and the average traveling speed per lane calculated in step 2c. Here, when it is determined that the traffic flow is "free flow", the process proceeds to step 2e, and when it is determined that "congestion" or "congestion", the process proceeds to step 2f.

In step 2e, a sudden event in the free flow is determined. If it is determined that a sudden event has occurred, the process proceeds to step 2g. On the other hand, if it is determined that a sudden event has not occurred. Moves to step 2a.

In step 2f, a sudden event is determined at the time of congestion or traffic congestion, and if it is determined that a sudden event has occurred, the process proceeds to step 2g, while if a sudden event has not occurred. If determined, step 2a
Move to

In step 2g, the operator is notified of the occurrence of the "sudden event" through the monitoring monitor M, and the process proceeds to step 2h. In step 2h, it is determined whether or not a request to end the sudden event detection process has been issued from the control device S. Here, if the end request has been issued, the process is ended. If there is no end request, the process proceeds to step 2a.

FIGS. 3 and 4 show the number of traffic per lane counted in step 2b, the average traveling speed per lane calculated in step 2c, and the lane change rate per lane for each monitoring section area. In particular, FIG.
FIG. 4 shows a state in which a “sudden event” occurs during a free flow in a section 0, and FIG. 4 shows a state in which a “sudden event” occurs in a section of the cameras 8 to 10 during congestion or traffic congestion.

First, when an "sudden event" occurs at the time of free flow as shown in FIG. 3, a section (each of the cameras 7, 11) before and after the section of the cameras 8 to 10 in which the "sudden event" occurs. 3), the lane change rate fluctuates significantly as shown in FIG. 3 and the traffic volume becomes more non-uniform between lanes in the same section. If such a condition is found, it is determined in step 2e that an "outbreak event" has occurred in the first lane.

On the other hand, when an "outbreak event" occurs at the time of congestion or traffic congestion as shown in FIG. 4, in the section of the cameras 8 to 10 where the "outbreak event" occurs, as shown in FIG. The average speed decreases in all lanes, and the lane change rate fluctuates remarkably around the section of cameras 8 to 10,
The non-uniformity of the traffic volume and the average speed between the section of the occurrence of the “sudden event” (section of camera 7-section of camera 8) and the end (section of camera 10-section of camera 11) increases. When such a state is found, in step 2f, the cameras 8 to
It is determined that the “sudden event” has occurred in section 10.

As described above, in the road monitoring system having the above configuration, when detecting an incident, the traffic volume, average speed, and lane change rate of each lane that does not require advanced image processing capability, Based on the traffic volume between lanes, the average speed, and the lane change rate, an unexpected event occurring on the road is detected from these non-uniformities.

Therefore, according to the road monitoring system having the above configuration, the occurrence position of the sudden event can be specified from the non-uniformity based on the traffic volume, the average speed, and the lane change rate between the lanes between the monitoring section areas. It is also possible to detect a sudden event in an unmonitored section area without requiring a high image processing capability.

The present invention is not limited to the above embodiment. It goes without saying that various modifications can be made without departing from the spirit of the present invention.

[0037]

As described above, according to the present invention, the number of vehicles passing through each monitoring section photographed by each surveillance camera is determined based on the images photographed by each surveillance camera, and then the number of vehicles passing through the surveillance camera is determined. A deviation in the number of passing vehicles (or average speed, lane change rate) between sections is obtained, and a section in which a sudden event has occurred is detected based on the deviation.

Therefore, according to the present invention, the number of passing vehicles (or average speed, lane change rate) for each monitoring section that does not require a high image processing capability is measured, and the number of passing vehicles between different monitoring sections (or By calculating the deviation of the average speed and the lane change rate, it is possible to specify the section where the sudden event has occurred.
That is, according to the present invention, even if an unexpected event occurs in an unmonitored section in which no monitoring camera is arranged, the above-mentioned unexpected event occurs based on the deviation of the number of vehicles (or average speed, lane change rate) in the preceding and following monitored sections. Since the occurrence of an event can be detected, it is possible to provide a road monitoring system capable of detecting a section where an unexpected event has occurred including an unmonitored section area without requiring a high image processing capability.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram showing a configuration of an embodiment of a road monitoring system according to the present invention.

FIG. 2 is a flowchart for explaining a sudden event detection process of the road monitoring system shown in FIG. 1;

FIG. 3 is a diagram showing fluctuations of various data including traffic volume obtained by the processing shown in FIG. 2;

FIG. 4 is a diagram showing changes in various data including traffic volume obtained by the processing shown in FIG. 2;

[Explanation of symbols]

 1 to N: surveillance camera V: image processing device S: control device M: surveillance monitor

Claims (8)

[Claims] A road surveillance system for arranging a plurality of surveillance cameras for photographing a road to be monitored and analyzing an image photographed by the surveillance camera to detect a sudden event occurring on the road. In the above, based on the video taken by each monitoring camera, vehicle number counting means for calculating the number of passing vehicles for each monitoring section captured by each monitoring camera, and the monitoring section determined by the vehicle number counting means Road surveillance, comprising: determining a deviation of the number of passing vehicles between different monitoring sections based on the number of passing vehicles for each, and detecting a section in which a sudden event has occurred based on the deviation. system. 2. An average speed measuring means for obtaining an average speed of a vehicle passing through each monitoring section photographed by each surveillance camera based on an image photographed by each surveillance camera, wherein the sudden event detecting means comprises: A deviation of an average speed between different monitoring sections is obtained based on the average speed for each monitoring section obtained by the average speed measuring means, and a section where an unexpected event has occurred is detected based on the deviation. The road monitoring system according to claim 1. 3. The number-of-vehicles counting means obtains the number of vehicles passing through each of the plurality of lanes of each monitoring section in each of the monitoring sections based on an image captured by a monitoring camera. Is the deviation of the number of vehicles passing between different monitoring sections based on the number of vehicles passing through a plurality of lanes in each monitoring section obtained by the vehicle number counting means, and the number of vehicles passing between lanes in the same monitoring section. The road monitoring system according to claim 1 or 2, wherein deviations are obtained, and a section where the sudden event has occurred is detected based on the deviations. 4. An average speed measurement unit calculates an average speed of vehicles passing through each of the plurality of lanes in each of the monitoring sections based on an image captured by a monitoring camera. The means includes a deviation of an average speed between different monitoring sections based on an average speed for each of a plurality of lanes of each monitoring section obtained by the average speed measuring means, and a deviation of an average speed between lanes in the same monitoring section. The road monitoring system according to claim 2 or 3, wherein a section in which an unexpected event has occurred is detected based on these deviations. 5. A lane change rate measuring means for obtaining, for each lane, a percentage of vehicles that change lanes for each monitoring section photographed by each surveillance camera, based on an image photographed by each surveillance camera, The sudden event detecting means obtains a deviation of a lane change rate between different monitoring sections based on the lane change rate obtained by the lane change rate measuring means, and detects a section where the sudden event has occurred based on the deviation. The road monitoring system according to any one of claims 1 to 4, wherein: 6. A road surveillance system for arranging a plurality of surveillance cameras for photographing a road to be monitored and analyzing an image photographed by the surveillance camera to detect a sudden event occurring on the road. An average speed measuring means for calculating an average speed of a vehicle passing through each monitoring section photographed by each surveillance camera based on an image photographed by each surveillance camera; and A road surveillance system comprising: a sudden event detection means for determining a deviation of an average speed between different monitoring sections based on an average speed of each section, and detecting a section where an unexpected event has occurred based on the deviation. . 7. The average event measuring means obtains an average speed of a vehicle passing through each of the plurality of lanes in each of the monitoring sections based on an image captured by a monitoring camera. The means includes a deviation of an average speed between different monitoring sections based on an average speed for each of a plurality of lanes of each monitoring section obtained by the average speed measuring means, and a deviation of an average speed between lanes in the same monitoring section. The road monitoring system according to claim 6, wherein the section in which the sudden event has occurred is detected based on these deviations. 8. A road surveillance system for arranging a plurality of surveillance cameras for photographing a road to be monitored and analyzing an image photographed by the surveillance camera to detect a sudden event occurring on the road. A lane change rate measuring means for obtaining, for each lane, a ratio of vehicles performing lane change, for each of the monitoring sections photographed by each surveillance camera, based on the video taken by each surveillance camera; Means for determining a deviation of a lane change rate between different monitoring sections based on the lane change rate obtained by the means, and detecting a section where an unexpected event has occurred based on the deviation based on the deviation. A road monitoring system characterized in that: