JP2007026301A - Stopping/low-speed vehicle detector and stopping/low-speed vehicle detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a stopping/low-speed vehicle detector and a stopping/low-speed vehicle detection method, capable of detecting only a traffic disturbance factor necessary for traffic control. <P>SOLUTION: A stopping/low-speed vehicle recognition part 2 recognizes, based on a photographic image of vehicles on a road, a stopping/low-speed vehicle which is in stopping state or low-speed state. A congested flow determination part 4 compares a traffic state quantity measured by a traffic state quantity measuring part 3 with a congestion determination threshold to determine whether the traffic flow state is a congested flow or not. A traffic flow abnormality determination part 3 releases, upon recognition of stopping/low-speed vehicle by the recognition part 2, the recognition of stopping/low-speed vehicle when the determination by the congested flow determination means shows the congested flow, and determines abnormality in traffic flow when the determination by the congested flow determination means shows no congested flow. A traffic flow abnormality detection part 6 detects traffic flow abnormality with the recognized stopping/low-speed vehicle as a sudden event when the abnormality in traffic flow is determined. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a stopped low-speed vehicle detection device and a stopped low-speed vehicle detection device method for performing processing based on a photographed image on a road and detecting a vehicle that has stopped or slowed down due to a failure or the like.

  Conventionally, a video of a vehicle traveling on a highway or a general road is photographed with a surveillance camera, and the vehicle image is displayed on a video display in a control room. Monitoring systems are widely adopted. In addition, an unexpected event is automatically detected by analyzing the video of the surveillance camera, the occurrence of the sudden event is displayed, and the incident video is switched to the on-site video of the sudden event occurrence point and displayed. Has been proposed (see, for example, Patent Document 1).

  Various techniques for detecting a vehicle from an image of a vehicle on a road have been proposed, and a vehicle type capable of determining a vehicle type at night with high accuracy by extracting and judging a headlight of the vehicle. Discriminating devices and methods have been proposed (see, for example, Patent Document 2).

  By detecting the vehicle from the captured image on the road as described above and tracking the detected vehicle in the captured image, it is possible to calculate the speed of the vehicle, detect a stopped vehicle, etc. It is possible to grasp the traffic situation in the area.

JP-A-5-250595 JP 11-353580 A

  However, if an unexpected event such as a vehicle that has stopped due to a malfunction or the like is considered as a traffic obstruction factor that is necessary for the traffic administration system, there are vehicles that are stopped or running at low speed due to traffic jams, When a change in illumination is erroneously detected as a vehicle, all are detected as a stopped low-speed vehicle. Therefore, there has been a situation in which the detection of a stopped low-speed vehicle frequently occurs due to a cause other than sudden events, that is, a cause that is not necessary for the traffic control system.

  The present invention has been made in view of the above-described conventional circumstances, and provides a stopped low-speed vehicle detection device and a stopped low-speed vehicle detection method capable of detecting only a traffic obstruction factor necessary for a traffic administration system. Objective.

  A stop low-speed vehicle detection device according to the present invention is based on a captured image obtained by shooting a vehicle on a road, a stop low-speed vehicle recognition means for recognizing a stop low-speed vehicle in a stop state or a low-speed state, and a vehicle that monitors a predetermined area A traffic state quantity measuring means connected to a detector for measuring the traffic state quantity in the area, and comparing the measured traffic state quantity and a congestion judgment threshold value, When the jammed flow determining means for judging whether or not the stopped low-speed vehicle is recognized, if the judgment by the jammed flow judging means is a jammed flow, the recognition of the stopped low-speed vehicle is canceled, and the jammed flow Traffic flow abnormality determining means for determining that there is an abnormality in the traffic flow when the determination by the determining means is not a traffic jam flow.

  With this configuration, the stopped low-speed vehicle in a traffic jam is excluded from the detection target of the traffic flow abnormality, so that it is possible to detect only the traffic obstruction factor necessary for the traffic administration system.

  Further, in the stopped low-speed vehicle detection device of the present invention, the traffic flow abnormality determining means determines that the traffic flow state of the region monitored by a vehicle detector provided downstream from the stopped low-speed vehicle is a traffic jam flow. The recognition of the stopped low-speed vehicle is canceled.

  With this configuration, if there is a traffic jam ahead of the recognized direction of travel of the slow low-speed vehicle, it is estimated that the recognized slow low-speed vehicle is also caused by the traffic jam. Since it is excluded from the detection target of the flow abnormality, it is possible to detect only the traffic obstruction factors necessary for the traffic administration system.

  In the stop low-speed vehicle detection device of the present invention, the traffic state quantity measuring unit measures the traffic volume from the number of vehicles passing through the monitoring area per unit time as the traffic state quantity, and the traffic congestion judging unit includes: When the traffic volume is greater than or equal to the traffic congestion determination threshold, it is determined that the traffic flow is a traffic jam.

  With this configuration, it is possible to determine a traffic jam based on the measured traffic volume and detect a stopped low-speed vehicle.

  Further, in the stop low-speed vehicle detection device according to the present invention, the traffic state quantity measuring unit measures an average speed of individual vehicles passing through the monitoring area as the traffic state quantity, and the traffic jam determining unit When the speed is equal to or less than the speed congestion determination threshold, it is determined that the flow is a congestion flow.

  With this configuration, it is possible to determine a traffic jam flow based on the measured average speed of the individual vehicle and detect a stopped low-speed vehicle.

  In the stop low-speed vehicle detection device according to the present invention, the vehicle detection device is an imaging device, and the traffic state quantity measuring unit changes a luminance value of a photographed image photographed by the photographing device as the traffic state quantity. The vehicle group speed is measured on the basis of the vehicle speed, and the traffic flow determining means determines that the traffic flow is a traffic jam when the vehicle group speed is equal to or less than a speed traffic congestion determination threshold value.

  With this configuration, because the vehicle group speed is used as the traffic state quantity, it is possible to detect a traffic flow abnormality with high accuracy even when it is difficult to extract the characteristics of the vehicle such as in a traffic jam.

  Further, in the stop low-speed vehicle detection device of the present invention, the vehicle detection device is an imaging device, and the traffic state quantity measuring unit is configured to occupy a space of the vehicle in a captured image taken by the imaging device as the traffic state quantity. The traffic flow determination means determines that the traffic flow is a traffic congestion flow when the occupation rate is equal to or greater than an occupation rate traffic congestion determination threshold.

  With this configuration, it is possible to determine a traffic jam flow based on the space occupancy rate of the vehicle in the captured image and detect a stopped low-speed vehicle.

  In the stop low-speed vehicle detection device of the present invention, the vehicle detector may be any one of an imaging device, an ultrasonic vehicle detector, a loop coil vehicle detector, and a tape vehicle detector. Composed. With this configuration, the traffic state quantity can be measured using various vehicle detectors.

  A stop low-speed vehicle detection device according to the present invention includes a stop low-speed vehicle recognition means for recognizing a stop low-speed vehicle in a stopped state or a low-speed state based on a captured image obtained by shooting a vehicle on a road, and the recognized stop low-speed vehicle. The low-speed holding information storage means for holding the information, and when the stopped low-speed vehicle is recognized, the information on the held low-speed vehicle is referred to, and the vehicle is different from the recognized stopped low-speed vehicle. Monitors whether or not the vehicle has passed through a monitoring area including the position of the stopped low-speed vehicle within a predetermined monitoring time, and if the passage of the other vehicle is detected, the recognition of the stopped low-speed vehicle is canceled. Traffic flow abnormality determining means for determining that there is an abnormality in the traffic flow when the passage of the other vehicle is not detected.

  With this configuration, an event that is erroneously detected due to sunlight or a change in lighting is excluded from a traffic flow abnormality detection target, so that only a traffic obstruction factor that is necessary for the traffic administration system can be detected.

  Further, in the stop low-speed vehicle detection device of the present invention, the stop low-speed holding information storage means stores a tail position of the stop low-speed vehicle as information on the stop low-speed vehicle, and the traffic flow abnormality determination means A vehicle passing through a stored vehicle rear position or a position upstream by a predetermined distance from the vehicle rear position is monitored.

  With this configuration, it is possible to detect only the traffic obstruction factors necessary for the traffic administration system by monitoring the vehicle tail position.

  Further, the stop low-speed vehicle detection device of the present invention is based on the captured image, a traffic state amount measuring means for measuring a traffic state amount in the area being photographed, and a traffic congestion determination with the measured traffic state amount. A traffic flow determination means for comparing the threshold value with the threshold value to determine whether or not the traffic flow state in the region is a traffic congestion flow, and the traffic flow abnormality determination means is adapted to determine the traffic flow according to the determination of the traffic congestion flow. Control the monitoring time.

  With this configuration, a stopped low-speed vehicle in a traffic jam is also excluded from the traffic flow abnormality detection target, so that only the traffic obstruction factors necessary for the traffic administration system can be detected.

  The stop low-speed vehicle detection method of the present invention includes a step of recognizing a stop low-speed vehicle in a stopped state or a low-speed state based on a captured image obtained by shooting a vehicle on a road, and a vehicle detector that monitors a predetermined area. Based on the detection signal, the step of measuring the traffic state quantity in the region is compared with the measured traffic state quantity and a congestion determination threshold value, and whether or not the traffic flow state in the area is a traffic jam flow. And when the stop low-speed vehicle is recognized, if the determination by the traffic flow determining means is a traffic jam flow, the recognition of the stopped low-speed vehicle is canceled and the determination by the traffic flow determining means is performed. A step of determining that there is an abnormality in the traffic flow when it is not a traffic jam flow.

  By this method, the stopped low-speed vehicle in a traffic jam is excluded from the detection target of the traffic flow abnormality, so that it is possible to detect only the traffic obstruction factor necessary for the traffic administration system.

  Further, the stop low-speed vehicle detection method of the present invention includes a step of recognizing a stop low-speed vehicle in a stop state or a low-speed state based on a captured image obtained by shooting a vehicle on a road, and information on the recognized stop low-speed vehicle. And when the stopped low-speed vehicle is recognized, a vehicle other than the recognized stopped low-speed vehicle is referred to by referring to the information on the held stopped low-speed vehicle. Is detected within a predetermined monitoring time, and when the passage of the other vehicle is detected, the recognition of the stopped low-speed vehicle is canceled and the passage of the other vehicle is detected. Otherwise, determining that there is an abnormality in the traffic flow.

  According to this method, an event erroneously detected due to sunlight or a change in illumination is excluded from a traffic flow abnormality detection target, so that only a traffic obstruction factor necessary for the traffic administration system can be detected.

  ADVANTAGE OF THE INVENTION According to this invention, the stop low-speed vehicle detection apparatus and stop low-speed vehicle detection method which can detect only the traffic obstruction factor required on a traffic administration system can be provided.

  In this embodiment, a road monitoring system that can detect a traffic flow abnormality by photographing a vehicle traveling with a camera installed in a tunnel or the like from behind, performing vehicle feature extraction from the captured image, vehicle recognition, etc. The example of a structure in is shown.

  FIG. 9 is an explanatory diagram showing an overview of a road monitoring system including a stopped low-speed vehicle detection device according to an embodiment of the present invention. As shown in FIG. 9, the road monitoring system is provided as an imaging device that is an example of a vehicle detector at each predetermined position on the roadside such as in a tunnel on the road 11, and a plurality of cameras 12 that monitor different areas. It has.

  The camera 12 captures a vehicle 13 traveling on the road 11 from behind and outputs a captured image signal SC as an example of a vehicle detection signal. In addition, the road monitoring system includes a stop low-speed vehicle detection device 10 that detects a vehicle, calculates a vehicle speed, detects a fallen object, and the like based on a captured image captured by the camera 12 to detect a traffic jam or suddenly. Perform event determination on the target road, such as event detection.

  In the present embodiment, the vehicle is photographed from the rear to prevent halation, smear, and the like due to the vehicle headlight. A vehicle is extracted based on the photographed image from the rear, and the vehicle width, the vehicle length, and the vehicle rear position are detected. Then, the detected vehicle is tracked, and when the tracking is impossible, the stop low-speed vehicle is detected by performing processing described later.

  Hereinafter, the stopped low-speed vehicle detection device 10a according to the first embodiment of the present invention and the stopped low-speed vehicle detection device 10b according to the second embodiment will be described.

(First embodiment)
FIG. 1 is a block diagram showing a schematic configuration of a stopped low-speed vehicle detection device according to the first embodiment of the present invention. As shown in FIG. 1, the stopped low-speed vehicle detection device 10 a according to the first embodiment is based on the output of the image processing / vehicle recognition unit 1, the stopped low-speed vehicle recognition unit 2, and the image processing / vehicle recognition unit 1. Traffic state quantity measuring unit 3 that measures the traffic state quantity, traffic flow judgment unit 4 that determines a traffic jam flow based on the output of the traffic state quantity measurement unit 3, output from the stop low speed vehicle recognition unit 2 and traffic jam flow judgment A traffic flow abnormality determination unit 5 that determines the presence or absence of a traffic flow abnormality based on the output of the unit 4, and a traffic flow abnormality detection unit 6 that detects a traffic flow abnormality based on the determination by the traffic flow abnormality determination unit 5.

  The image processing / vehicle recognition unit 1 performs image processing based on the captured image signal SC input from the camera 12 and performs vehicle recognition based on the processing result. The stopped low-speed vehicle recognition unit 2 recognizes a stopped low-speed vehicle in a stopped state or a low-speed state based on the vehicle information RC recognized by the image processing / vehicle recognition unit 1.

  The traffic state quantity measuring unit 3 includes a traffic volume calculating unit 3a, a speed calculating unit 3b, and an occupation rate calculating unit 3c. Based on the image processing information and vehicle information from the image processing / vehicle recognition unit 1, The traffic state quantity in the imaging region of the camera 12 is measured.

  Then, as the traffic state quantity, the traffic volume calculation unit 3a measures the number of vehicles passing through the monitoring area of the camera 12 per unit time, and the speed calculation unit 3b calculates the average speed of the individual vehicles passing through the monitoring area of the camera 12 and The vehicle group speed is measured, and the occupancy rate calculation unit 3 c measures the space occupancy rate of the vehicle in the monitoring area of the camera 12. In addition, you may employ | adopt the average value in predetermined time, such as the past 1 minute, for these traffic state quantities, for example.

  The traffic flow judgment unit 4 compares the traffic state quantity measured by the traffic state quantity measurement unit 3 with the traffic judgment threshold value and judges whether or not the traffic flow in the monitoring area of the camera 12 is a traffic jam flow. When the stop low-speed vehicle recognition unit 2 recognizes the stopped low-speed vehicle, the traffic flow abnormality determination unit 5 cancels the recognition of the stopped low-speed vehicle when the determination by the traffic jam flow determination unit 4 is a traffic jam flow. When the determination by the determination unit 4 is not a traffic jam, it is determined that there is an abnormality in the traffic flow. When it is determined that there is an abnormality in the traffic flow, the traffic flow abnormality detection unit 6 detects a traffic flow abnormality in which the recognized stopped low-speed vehicle is a sudden event.

  FIG. 2 is a flowchart showing a processing procedure of image processing according to the first embodiment of the present invention, and outlines operations of the image processing / vehicle recognition unit 1, the stopped low-speed vehicle recognition unit 2, and the traffic state quantity measurement unit 3. Is described.

  The image processing / vehicle recognition unit 1 first creates background image data and background differential image data as initial data (step S201). Here, after power-on or after resetting, for example, 300 times of captured image data (images for 30 seconds) taken from the camera 12 every 0.1 seconds are averaged to obtain a background image. Also, a background differential image is created by performing differential processing of the background image data. Furthermore, since the brightness of the illumination on the road is different between night and day, the night mode or the day mode is determined based on the background image, and the mode corresponding to the current time is set.

  Next, image data used for image analysis is created from the captured image data captured from the camera 12 (step S202). Here, the video signal based on the NTSC signal input from the camera 12 has a vertical direction of 240 lines, a horizontal direction of 320 pixels, a luminance of 256 gradations, and is subjected to AD conversion (analog-digital conversion) for each frame to 320 pixels × 240. Create pixel image data. At this time, only an even field is AD-converted to create one image data every 1/30 sec. In the subsequent image analysis, three image data of the current image, the image before ΔT, and the image before 2ΔT (here, ΔT = 100 ms) are used.

  Then, image preprocessing such as feature extraction calculation and binarization processing is performed on the created image data to create a feature image from which the features of the vehicle are extracted (step S203). In the feature extraction calculation, a difference image obtained by performing feature extraction using a feature extraction method such as a background difference method, a differential background difference method, a frame difference method, or a differential frame difference method is created, and a feature image is created. Here, in order to increase the processing speed of the post-processing, for example, a feature image is created by performing processing such as feature extraction, and then the image data of 320 pixels × 240 pixels is compressed to 80 pixels × 60 pixels. Create an extracted compressed image. Further, a light extraction image of 320 pixels × 240 pixels is created from the image data created for image analysis.

  Subsequently, image analysis is performed using the feature extraction compressed image and the light extraction image, and a vehicle or the like is extracted (step S204). At this time, the vehicle body analysis is performed using the feature extraction compressed image, and the stern, the vehicle length, and the vehicle width are determined. Further, light analysis is performed using the light extraction image, and the vehicle tail is determined by detecting the light of the vehicle. The vehicle position is determined from the results of the rectangular vehicle body detection by the vehicle body analysis and the vehicle tail detection by the light analysis. Further, the falling object analysis is performed using the feature extraction compressed image, and the position and the vertical and horizontal dimensions of the falling object are determined.

  After that, the detected vehicle position data is added or newly created to the travel locus data, and the vehicle tracking is performed by adding the detected vehicle position data to the travel locus data for each cycle of this processing (step S205). Then, the traffic state quantity measuring unit 3 measures the traffic state quantity. Specifically, the traffic volume calculation unit 3a counts one vehicle when the vehicle tail detected in step S204 passes a predetermined horizontal line of the image, and determines the number of vehicles passing per predetermined time as traffic. Calculate as a quantity. Further, the speed calculation unit 3b calculates the detected moving speed of each vehicle. Furthermore, the occupation rate calculation unit 3c calculates, as an occupation rate, the proportion of the vehicle in one screen of the image captured by the camera 12 from the vehicles extracted in step S204.

  Further, the speed calculation unit 3b of the traffic state quantity measurement unit 3 calculates the speed of the vehicle group in the captured image by a method different from the vehicle tracking (step S206). Here, the entire vehicle is regarded as one lump car group from the movement amount of the luminance value of the entire image, and the moving speed of this car group is calculated. This vehicle group speed can be used for determining a traffic jam at the time of a traffic jam.

  Next, the stop low speed vehicle recognition unit 2 recognizes whether or not an event such as “stop” or “low speed” has occurred from the results of the above-described vehicle tracking, vehicle group speed calculation, falling object detection, and the like (step S207). . In addition, the stop low-speed vehicle recognition part 2 provides an identification number as one of vehicle information for every extracted vehicle body. A case where the stern of the vehicle body with the same identification number does not move for more than n seconds is recognized as a stopped state, and a case where the speed of an individual vehicle with the same identification number continues for less than m [km] for 1 second is recognized as a low speed state. Note that n, m, and l may be set variably according to time and the like.

  Then, the image processing / vehicle recognition unit 1 updates the background image data and the background differential image data to the latest background image for image processing in the next processing cycle (step S208). Further, the night mode or the day mode is determined based on the new background image, and the mode corresponding to the current time is updated. Then, it returns to step S202 and repeats the processing cycle of steps S202-S208 for every predetermined interval (here 100 ms).

  FIG. 3 is a block diagram showing a schematic configuration of the image processing / vehicle recognition unit according to the first embodiment of the present invention.

  The image processing / vehicle recognition unit 1 includes an image input unit 21 that inputs image data captured by the camera 12, a background image generation unit 22 that generates a background image from the input image data, a background difference using the input image and the background image, A feature image creation unit 23 that creates a feature image by performing processing such as differential background difference, frame difference, and differential frame difference, and creates a feature code indicating the feature of each pixel for calculating a vehicle candidate rectangle based on the feature image A feature code creation unit 24, an image storage unit 25 that stores image data, a feature code storage unit 29 that stores a feature code, and a vehicle recognition unit 30 that performs vehicle recognition processing.

  The image storage unit 25 includes a current image storage unit 26 that stores a current image, a background image storage unit 27 that stores a background image, and a feature image storage unit 28 that stores a feature image. The vehicle recognizing unit 30 includes a vehicle head / vehicle tail detection unit 31 that detects the vehicle head or the vehicle rear from the current image and the calculated feature code, and a vehicle tracking unit 32 that tracks the detected vehicle using the vehicle head / car tail. The image storage unit 25 and the feature code storage unit 29 are configured by a memory or the like. The background image creation unit 22, the feature image creation unit 23, the feature code creation unit 24, and the vehicle recognition unit 30 are configured by a processor or the like that performs each process.

  The current image storage unit 26 stores the current image data input from the image input unit 21. As input image data, a grayscale image of 256 gradations is used. The background image creation unit 22 creates a background image of only the background using past current image data and stores the background image in the background image storage unit 27. The feature image creation unit 23 uses the current image data stored in the current image storage unit 26 and the background image data stored in the background image storage unit 27 to use a background difference, a differential background difference, a frame difference, and a differential frame difference. The feature image for detecting the vehicle in the photographed image is created and stored in the feature image storage unit 28.

  The feature code creation unit 24 creates a feature code indicating the feature of each pixel for calculating the vehicle candidate rectangle based on the feature image, and stores the feature code in the feature code storage unit 29. The feature code is obtained by assigning codes such as 0, 1, 2,..., 9 for each pixel in the feature image. For example, 0 is the part that is the same as the background and no data is obtained by differential processing, 1 is the discarded data that seems to be noise, etc., 2 is the effective data valid as a vehicle, 2 is the part that is smaller than the vehicle width, Is data that makes it possible to determine the position of the vehicle, such that 3 is a candidate vehicle part, 8 is a part recognized as a vehicle body, 9 is a part recognized as a car tail, and so on.

  The vehicle head / vehicle tail detection unit 31 calculates the position of the vehicle head or the vehicle rear using the current image data stored in the current image storage unit 26 and the feature code stored in the feature code storage unit 29. The vehicle tracking unit 32 traces the position of the detected vehicle by using the position of the head or tail of the vehicle obtained in time series by the above processing to calculate a travel locus, or calculates the moving speed of the detected vehicle. Do tracking. The vehicle recognition unit 30 may detect a vehicle candidate rectangle using the current image data and the feature code, and perform vehicle recognition using the vehicle candidate rectangle. In this case, the vehicle can be recognized even when the vehicle head or the vehicle tail is outside the screen.

  FIG. 4 is a flowchart showing a processing procedure for stop low-speed vehicle detection according to the first embodiment of the present invention. The traffic flow abnormality determination unit 5 monitors the presence or absence of the stopped low-speed vehicle until the stopped low-speed vehicle recognition unit 2 recognizes the stopped low-speed vehicle (step S401). When the stopped low-speed vehicle recognition unit 2 recognizes the stopped low-speed vehicle (YES in step S401), the result of the traffic jam flow determination process (step S402) by the traffic jam flow determination unit 4 is acquired.

  FIG. 5 is a flowchart showing a procedure for determining a traffic jam according to the first embodiment of the present invention. The traffic jam determination unit 4 acquires the traffic volume, speed, and occupation rate, which are information on the traffic volume measured by the traffic volume measuring unit 3 (step S501), and determines whether the traffic volume is equal to or greater than a threshold value. Determination is made (step S502). If the traffic volume is equal to or greater than the threshold value (YES in step S502), it is determined whether the speed is equal to or less than the threshold value (step S503). For example, when the vehicle group speed is used for this speed determination, it is possible to accurately measure the speed even when it is difficult to extract the characteristics of the vehicle such as in a traffic jam. It is possible to detect abnormalities in traffic flow with high accuracy.

  If the speed is equal to or lower than the threshold value (YES in step S503), it is determined whether or not the occupation ratio is equal to or higher than the threshold value (step S504). If the occupation ratio is equal to or greater than the threshold value (YES in step S504), it is determined that the traffic flow state is a traffic jam flow (step S505).

  On the other hand, if the traffic volume is less than the threshold (NO in step S502), the speed is greater than the threshold (NO in step S503), and the occupation rate is less than the threshold (NO in step S504), the traffic It is determined that the flow state is a free flow.

  In the above description, the example of determining the traffic flow state by combining all of the traffic volume, the speed, and the occupation ratio as the traffic state amount has been described. However, only one of these or two traffic state amounts are described. May be used to determine the traffic flow state. Moreover, as a traffic state quantity, it is not restricted to traffic volume, speed, and an occupation rate, If a value which shows a traffic state, various state quantities can be applied.

  Returning to FIG. 4, the traffic flow abnormality determination unit 5 acquires the determination by the traffic congestion flow determination unit 4, and the traffic flow state of the area monitored by the camera 12 provided downstream from the recognized low-speed vehicle is traffic congestion. If it is determined that the current is a flow (YES in step S403), it is determined that there is an abnormality in the traffic flow. Then, the traffic flow abnormality detection unit 6 detects a traffic flow abnormality (step S404). On the other hand, when it is determined that the traffic flow state in the area monitored by the camera 12 provided downstream from the recognized stopped low-speed vehicle is not a traffic jam (NO in step S403), the stopped low-speed vehicle recognition unit 2 stops the low speed. Cancel vehicle recognition.

  According to the first embodiment of the present invention as described above, the stopped low-speed vehicle in a traffic jam is excluded from the detection target of the traffic flow abnormality, so that it is possible to detect only the traffic obstruction factor necessary for the traffic administration system. Further, by using the congestion flow determination of the area monitored by the camera 12 provided downstream from the recognized stopped low-speed vehicle, when the traffic congestion has occurred in the traveling direction of the recognized stopped low-speed vehicle, Because it is estimated that the stopped low-speed vehicle is also caused by traffic congestion, the stopped low-speed vehicle in the traffic congestion is excluded from the detection target of the traffic flow abnormality, so it is possible to detect only the traffic obstruction factors necessary for the traffic administration system. .

(Second Embodiment)
FIG. 6 is a block diagram showing a schematic configuration of a stopped low-speed vehicle detection device according to the second embodiment of the present invention. In the figure, the same reference numerals are given to the portions overlapping those in FIG. 1 described in the first embodiment.

  The stop low-speed vehicle detection device 10b according to the second embodiment includes a stop low-speed holding information storage unit 7 that holds information on a stopped low-speed vehicle recognized by the stop low-speed vehicle recognition unit 2, and the traffic flow abnormality determination unit 8 includes: The traffic flow abnormality is determined based on the recognition result by the stop low-speed vehicle recognition unit 2, the information held in the stop low-speed holding information storage unit 7, and the real-time vehicle recognition result from the image processing / vehicle recognition unit 1.

  FIG. 7 is a flowchart showing a processing procedure for stop low-speed vehicle detection according to the second embodiment of the present invention.

  The traffic flow abnormality determination unit 8 monitors the presence or absence of the stopped low-speed vehicle until the stopped low-speed vehicle recognition unit 2 recognizes the stopped low-speed vehicle (step S701). When the stopped low-speed vehicle recognition unit 2 recognizes the stopped low-speed vehicle (YES in step S701), the traffic flow abnormality determination unit 8 recognizes the stopped low-speed vehicle, and the stopped low-speed holding information storage unit 7 The vehicle information output from the recognition part 2 is stored (step S702). Examples of stored vehicle information are a recognized vehicle identification number and vehicle tail position information.

  The traffic flow abnormality determination unit 8 refers to the information about the stopped low-speed vehicle held in the stopped low-speed holding information storage unit 7, and based on the real-time vehicle recognition result output from the image processing / vehicle recognition unit 1, It is monitored whether a vehicle other than the stopped low-speed vehicle has passed through the monitoring area including the position of the stopped low-speed vehicle within the holding time that is the monitoring time (step S703).

  The traffic flow abnormality determination unit 8 is a vehicle rear position as vehicle information stored in the stop low speed holding information storage unit 7 or a position (s [m]) upstream from the vehicle rear position (from the rear of the vehicle). A vehicle passing through the monitoring area is monitored with the monitoring area as the image front side of the camera 12 to be photographed.

  The holding time is set according to the position of the camera monitored by the stopped low-speed vehicle. For example, in a place where erroneous detection of a vehicle due to sunlight or lighting is likely to occur, it is possible to prevent erroneous detection of a stopped low-speed vehicle by extending the holding time. In addition, you may set automatically according to time etc.

  If the passage of another vehicle is detected in the monitoring area (YES in step S703), the recognition of the stopped low-speed vehicle is canceled (step S704). On the other hand, when the passage of another vehicle is not detected in the monitoring area (NO in step S703), it is determined that the traffic flow is abnormal, and the traffic flow abnormality detection unit 6 detects the traffic flow abnormality (step S705). .

  That is, if the vehicle is in a stopped state or a low speed state, the position of the stopped low-speed vehicle (or a few meters before the tail position, etc.) and the following vehicle should not pass, so it is held at the recognized stopped low-speed vehicle position. When a vehicle passing in time is detected, it is possible to prevent erroneous detection of a traffic flow abnormality due to erroneous recognition of the stopped low-speed vehicle by canceling the recognition of the stopped low-speed vehicle.

  Note that the above-described holding time may be variably controlled according to the determination of the traffic jam flow. FIG. 8 is a flowchart showing a processing procedure for changing the holding time according to the second embodiment of the present invention.

  As shown in FIG. 8, the congestion flow determination unit 4 acquires the traffic state quantity at the stop low-speed vehicle detection point measured by the traffic state quantity measurement unit 3 (step S801), and based on the traffic state quantity. Judge the traffic flow. When the traffic flow abnormality determination unit 5 determines that the traffic flow state is a traffic jam flow (YES in step S802), the traffic flow abnormality determination unit 5 sets the holding time to t1 seconds (step 803). If it is determined that the traffic flow state is not a traffic jam (NO in step S802), the holding time is set to t2 seconds shorter than t1 seconds.

  In this way, by setting a longer holding time when there is a traffic jam, the possibility of canceling the slow low-speed vehicle recognition by the vehicle in the traffic jam is increased, so the stop low-speed vehicle detection with higher accuracy is performed. be able to.

  The holding time may be set every time the stopped low-speed vehicle recognition unit 2 recognizes the stopped low-speed vehicle, or may be automatically performed every predetermined time.

  According to the second embodiment of the present invention as described above, an event erroneously detected due to sunlight or a change in illumination is excluded from a traffic flow abnormality detection target, so that only a traffic obstruction factor necessary for the traffic administration system is detected. Can do.

  In the embodiment of the present invention, the case where the camera is applied to the vehicle detector for measuring the traffic state quantity has been described. However, the present invention is not limited to this, and the ultrasonic vehicle detector is a loop embedded under the road. A loop coil type vehicle detector that detects the vehicle using the electromotive force generated in the coil, or is provided on the road or in the ground, and when the vehicle passes, the vehicle is detected by connecting the switch by its weight. A tape-type vehicle detector or the like may be applied.

  The present invention has the effect of being able to detect only the traffic obstruction factors necessary for the traffic administration system, performs processing based on the photographed image on the road, and detects a vehicle that has stopped or slowed down due to a failure or the like This is useful for a stopped low-speed vehicle detection device, a stopped low-speed vehicle detection method, and the like.

The block diagram which shows schematic structure of the stop low-speed vehicle detection apparatus which concerns on the 1st Embodiment of this invention. 6 is a flowchart showing a processing procedure of image processing according to the first embodiment of the present invention. 1 is a block diagram showing a schematic configuration of an image processing / vehicle recognition unit according to a first embodiment of the present invention. The flowchart which shows the process sequence of the stop low-speed vehicle detection which concerns on the 1st Embodiment of this invention. The flowchart which shows the process sequence of the traffic congestion determination which concerns on the 1st Embodiment of this invention. The block diagram which shows schematic structure of the stop low-speed vehicle detection apparatus which concerns on the 2nd Embodiment of this invention. The flowchart which shows the process sequence of the stop low-speed vehicle detection which concerns on the 2nd Embodiment of this invention. The flowchart which shows the process sequence of the holding time change which concerns on the 2nd Embodiment of this invention. Explanatory drawing which shows the outline | summary of the road monitoring system containing the stop low-speed vehicle detection apparatus which concerns on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing / vehicle recognition part 2 Stop low-speed vehicle recognition part 3 Traffic state quantity measurement part 3a Traffic volume calculation part 3b Speed calculation part 3c Occupancy rate calculation part 4 Congested flow judgment part 5, 8 Traffic flow abnormality judgment part 6 Traffic flow abnormality Detection unit 7 Stop low-speed holding information storage unit 10, 10a, 10b Stop low-speed vehicle detection device 11 Road 12 Camera 13 Vehicle 21 Image input unit 22 Background image creation unit 23 Feature image creation unit 24 Feature code creation unit 25 Image storage unit 26 Current Image storage unit 27 Background image storage unit 28 Feature image storage unit 29 Feature code storage unit 30 Vehicle recognition unit 31 Vehicle candidate rectangle detection unit 32 Vehicle tracking unit

Claims (12)

A stopped low-speed vehicle recognition means for recognizing a stopped low-speed vehicle in a stopped state or a low-speed state based on a captured image of a vehicle on the road;
A traffic state quantity measuring means connected to a vehicle detector for monitoring a predetermined area and measuring a traffic state quantity in the area;
A traffic flow judgment means for comparing the measured traffic state quantity with a traffic judgment threshold and judging whether the traffic flow state in the area is a traffic jam flow;
When the stop low-speed vehicle is recognized, if the determination by the traffic flow determining unit is a traffic jam flow, the recognition of the stopped low-speed vehicle is canceled, and if the determination by the traffic jam flow determining unit is not a traffic jam flow, traffic A traffic flow abnormality determining means for determining that there is an abnormality in the flow;
A low-speed vehicle detection device comprising: The stop low-speed vehicle detection device according to claim 1,
The traffic flow abnormality determination means cancels the recognition of the stopped low-speed vehicle when it is determined that the traffic flow state of the area monitored by the vehicle detector provided downstream from the stopped low-speed vehicle is a traffic jam flow. Stop low-speed vehicle detection device. The stop low-speed vehicle detection device according to claim 1 or 2,
The traffic state quantity measuring means measures the traffic volume from the number of vehicles passing through the monitoring area per unit time as the traffic state quantity,
The stop low-speed vehicle detection device that determines that the traffic is a traffic jam when the traffic volume is equal to or greater than a traffic jam determination threshold. A stop low-speed vehicle detection device according to any one of claims 1 to 3,
The traffic state quantity measuring means measures an average speed of individual vehicles passing through the monitoring area as the traffic state quantity,
The low-speed vehicle detection device that determines that the traffic flow is a traffic flow when the average speed is equal to or less than a speed traffic determination threshold. The stop low-speed vehicle detection device according to any one of claims 1 to 4,
The vehicle detection device is an imaging device,
The traffic state quantity measuring means measures a vehicle group speed as the traffic state quantity based on a change in luminance value of a photographed image photographed by the photographing device,
The stop traffic low speed vehicle detection device that determines that the traffic flow is a traffic jam when the vehicle group speed is equal to or less than a speed traffic congestion determination threshold. A stop low-speed vehicle detection device according to any one of claims 1 to 5,
The vehicle detection device is an imaging device,
The traffic state quantity measuring means measures the space occupancy rate of the vehicle in the captured image taken by the imaging device as the traffic state quantity,
The low-speed vehicle detection device that determines that the traffic flow is a traffic flow when the occupation rate is equal to or greater than an occupation rate traffic determination threshold. The stop low-speed vehicle detection device according to any one of claims 1 to 4,
The vehicle detector is a stopped low-speed vehicle detector configured by any one of an imaging device, an ultrasonic vehicle detector, a loop coil vehicle detector, and a tape vehicle detector. A stopped low-speed vehicle recognition means for recognizing a stopped low-speed vehicle in a stopped state or a low-speed state based on a captured image of a vehicle on the road;
Stop low speed holding information storage means for holding information of the recognized stopped low speed vehicle;
When the stopped low-speed vehicle is recognized, a monitoring area including a position of the stopped low-speed vehicle is referred to by referring to the information on the held stopped low-speed vehicle. , Monitoring whether or not the vehicle has passed within a predetermined monitoring time, and if the passage of the other vehicle is detected, the recognition of the stopped low-speed vehicle is canceled and the passage of the other vehicle is not detected. A traffic flow abnormality determining means for determining that there is an abnormality in the traffic flow;
A low-speed vehicle detection device comprising: The stop low-speed vehicle detection device according to claim 8,
The stop low speed holding information storage means stores the tail position of the stop low speed vehicle as information on the stop low speed vehicle,
The said traffic flow abnormality determination means is a stop low-speed vehicle detection apparatus which monitors the vehicle which passes the position of a predetermined distance upstream from the said vehicle rear position or the said vehicle rear position. The stop low-speed vehicle detection device according to claim 8 or 9,
Based on the photographed image, the traffic state quantity measuring means for measuring the traffic state quantity in the area being photographed is compared with the measured traffic state quantity and a congestion determination threshold value, and traffic in the area is represented. A traffic flow judgment means for judging whether the flow state is a traffic jam flow,
The said traffic flow abnormality determination means is a stop low-speed vehicle detection apparatus which controls the said monitoring time according to determination of the said traffic jam flow. Recognizing a stopped low-speed vehicle in a stopped state or a low-speed state based on a captured image of a vehicle on the road; and
Measuring a traffic state quantity in the area based on a detection signal from a vehicle detector that monitors the predetermined area;
Comparing the measured traffic state quantity with a congestion determination threshold value to determine whether the traffic flow state in the area is a congestion flow; and
When the stop low-speed vehicle is recognized, if the determination by the traffic flow determining unit is a traffic jam flow, the recognition of the stopped low-speed vehicle is canceled, and if the determination by the traffic jam flow determining unit is not a traffic jam flow, traffic Determining that there is an abnormality in the flow;
A stop low-speed vehicle detection method comprising: Recognizing a stopped low-speed vehicle in a stopped state or a low-speed state based on a captured image of a vehicle on the road; and
Holding information of the recognized stopped low-speed vehicle;
When the stopped low-speed vehicle is recognized, a monitoring area including a position of the stopped low-speed vehicle is determined in advance by referring to the held information on the stopped low-speed vehicle. Whether or not the vehicle has passed within the monitoring time of the vehicle, and when the passage of the other vehicle is detected, the recognition of the stopped low-speed vehicle is canceled, and when the passage of the other vehicle is not detected, the traffic flow is canceled. Determining that there is an abnormality,
A stop low-speed vehicle detection method comprising:

Images