JPH0944788A - Vehicle sensor and vehicle sensing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To exactly recognize the locations of two vehicles at a point of time when the two vehicles are acquired as the image of a vehicle. SOLUTION: The main processor 22 detects the location of the same vehicle from the image of which input screen is binarized, and tracks the location. In a local memory 23, the detected location is stored. When the main processor 22 detects that two vehicles are acquired as the image of a vehicle by the change of the type of a vehicle at a moving expected location, the distance between the two vehicles at a point of previous time is determined and the location of the vehicle which becomes impossible to be detected at a point of present time is determined from the location of the vehicle at the point of present time and the distance between the vehicles.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle detector for detecting a running vehicle by recognizing a vehicle image in an image taken by an image pickup device such as a CCD camera, and a vehicle detection method thereof.

[0002]

2. Description of the Related Art For the purpose of measuring the flow of traffic, a vehicle detector for detecting a running vehicle has been developed by using image processing technology. Conventionally, in this type of device, a CCD camera is used to capture an image of a running vehicle. As shown in FIG. 5, an image showing a characteristic point of the vehicle (front end or rear end of the vehicle) and its position are detected from the photographing screen obtained in a constant cycle. Next, the characteristic points of the vehicle are tracked over time on the screen obtained a predetermined number of times. The time when the vehicle first enters the measurement area of the CCD camera in advance (time t in FIG. 5), the detection position of the characteristic point at that time, the time when the vehicle leaves the measurement area (time t + n in FIG. 5), the characteristic point at that time It is recognized that the vehicle has passed the measurement area from the detection position of. By repeating this process, the vehicle detector counts the number of vehicles passing through the measurement area and calculates parameters for measuring speed.

[0003]

As described above, in the prior art, in order to confirm the passage of a vehicle, the characteristic point of the vehicle that has entered the measurement area is present in the photographing screen photographed a predetermined number of times thereafter. The requirement is to be able to track the vehicle. However, if the following vehicle of a certain specific vehicle is a large vehicle, the characteristic points of the specific vehicle are hidden by the characteristic points of the large vehicle in the shooting screen. As a result, if a predetermined number of shooting screens cannot be obtained at the time when the characteristic points are hidden, the characteristic points of the vehicle cannot be tracked and the passage of the vehicle cannot be recognized. Therefore, the number of vehicles cannot be counted. In addition, when the photographing screens for the predetermined number of times are obtained, the vehicle detector erroneously determines that the vehicle has exited, even though the vehicle exists in the measurement area. When this situation occurs in speed measurement, the measured speed value of the vehicle becomes higher than the actual speed value. This causes a measurement error.

It should be noted that the image of one vehicle acquired from the photographing screen is divided, and if the divided images include the image features of the front end portion and the rear end portion showing the features of the vehicle, two images are obtained. Vehicle detectors have been proposed (OMRON TECHNICS Vol No. 2) in which it is determined that two vehicles are close to each other and image separation is performed to obtain two vehicle images.
(Vol. 102). However, this proposal is effective when the speed of the adjacent vehicle is low, but since image separation processing is performed on each vehicle image, it is necessary to perform the same processing as detecting a vehicle several times the actual vehicle. It will take a long time to finally obtain the speed, etc., and the images to be divided will differ depending on the large and small vehicle types, so if you make a mistake in distinguishing the vehicle type, the vehicle's sensing accuracy will decrease It cannot be said that it is suitable for vehicle detection, and there are still points to be improved.

Therefore, in view of the above points, the present invention has an object to provide a vehicle detector and a vehicle detection method thereof that improve the vehicle detection performance when vehicles are adjacent to each other.

[0006]

According to a first aspect of the present invention, there is provided a vehicle detector for photographing a vehicle that has entered a measurement area and measuring the behavior of the vehicle from the image position of the vehicle image on the photographing screen. Storage means for storing the position of the same vehicle obtained in series as history data for vehicle tracking,
Abnormality detection means for detecting an abnormal situation in which two vehicles traveling in the measurement region are close to each other and are photographed as one vehicle image in the photographing screen, and when the abnormal situation is detected, an abnormality is detected. One of the two vehicles, further comprising arithmetic processing means for correcting the vehicle position of one of the two vehicles at the present time based on the history data of the storage means before the present time when the situation was detected. The history data at the present time is created by regarding the position as the corrected vehicle position and the other vehicle position as the vehicle position obtained from the one vehicle image.

According to a second aspect of the present invention, in addition to the first aspect, the one vehicle position is a vehicle position of a preceding vehicle in the two vehicles, and the other vehicle position is the two vehicles position. Is the vehicle position of the following vehicle in the vehicle.

According to a third aspect of the present invention, the arithmetic processing means of the first aspect obtains the inter-vehicle distance of the two vehicles from the history data of the first aspect and the vehicle position of the succeeding vehicle of the second aspect. The vehicle position of the preceding vehicle of claim 2 is obtained from the vehicle position of the following vehicle of item 2 and the obtained inter-vehicle distance.

According to a fourth aspect of the invention, in addition to the first aspect of the invention, the arithmetic processing means calculates the speed of the vehicle using the history data and the photographing interval time.

According to a fifth aspect of the present invention, in a vehicle detector for photographing a vehicle that has entered a measurement area and measuring the behavior of the vehicle from the image position of the vehicle image in the photographing screen, two vehicles are used for the photographing screen. In the vehicle detection method when captured as one vehicle image in the vehicle, the position of the same vehicle obtained in time series from a plurality of photographing screens is stored in the vehicle detector as history data for vehicle tracking. An abnormal situation is detected in which two vehicles traveling in the measurement area approach each other and are photographed as one vehicle image in the photographing screen.
When the abnormal situation is detected, the one vehicle image is regarded as an image of the following vehicle of the two vehicles, and the current vehicle position of the following vehicle is acquired to obtain the two vehicles. Is acquired from the history data, and the current vehicle position of the preceding vehicle is acquired from the current vehicle position of the subsequent vehicle and the acquired inter-vehicle distance.

In the present invention, attention is paid to the fact that the distance between adjacent traveling vehicles is substantially constant for a short time, and in the inventions of claims 1 and 5, two vehicles are captured as one vehicle image. An abnormal situation that has occurred is detected, and this one vehicle image is regarded as one of the two vehicle images. Further, the other vehicle image is calculated from the inter-vehicle distance at the previous time point and the current vehicle position.

In the second and third aspects of the present invention, a small vehicle is often hidden by a large vehicle following it when it is about to leave the measurement area. The position of the vehicle is acquired using historical data, for example, the inter-vehicle distance.

According to the fourth aspect of the invention, the speed is obtained from the vehicle position and the photographing interval time, and not only the vehicle position but also the speed of the adjacent vehicle is acquired.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows the arrangement of vehicle detectors. The vehicle detector comprises a CCD camera 10 and a vehicle detector body 20. The CCD camera 10 is installed at a high place at the end of the road,
Take a picture of a vehicle traveling in multiple lanes. There are two CCD camera orientations, one for shooting the vehicle from the front and the other for shooting the vehicle from the rear. The vehicle detector body 20 is CC
By performing image processing on an image captured by the D camera 10 as described later, the adjacent vehicle can sense the vehicle.

FIG. 2 shows the system configuration of the vehicle detector.
In FIG. 2, the vehicle detector body 20 is a data input / output unit 1
1 and the image processing unit 21. CCD camera 10
The image for one screen sent from the A / D converter 12 is analog / digital converted and stored in the image memory 26. The preprocessor 28 dedicated to image processing mainly executes pixel processing, performs background image processing for extracting the entire object, spatial difference for obtaining the contour of the object, and other similar image processing in the related art to perform preprocessing of the captured image. To process. Various data used in this image processing are temporarily stored in the local memory 27.

Vehicle characteristic data obtained as a result of the above image processing is stored in the dual port memory 25 via the image bus 29. The main processor 22 reads out these vehicle characteristic data from the dual port memory 25, determines the vehicle type of the vehicle by a method such as pattern matching, detects the traveling position of the vehicle at that time, and detects the vehicle speed and the passing vehicle. Measure the number of. The processing executed by the preprocessor 28 and the main processor 22 will be described later with reference to FIG. The local memory 23 temporarily stores various data used in the processing of the main processor 22. The data of the speed and the number of passing vehicles obtained as the measurement result are included in the CPU bus 24, the image input / output port (I
/ O) 13 and is transmitted to the outside.

FIG. 3 shows the measurement process executed in such a configuration. The system operation of the vehicle detector and the vehicle detection method according to the present invention will be described with reference to FIG. The road has one lane for ease of explanation. Further, it is assumed that the measurement area (measurement sample point) used for tracking the vehicle is set on the local memory 23 prior to the measurement (step S10 in FIG. 3).

When a vehicle enters the measurement area, the vehicle moves to C
It is captured by the CD camera 10. The preprocessor 28 is C
An image for one screen taken by the CD camera 10 is read from the image memory 26, and pre-processing for vehicle detection is executed. More specifically, the image of the approaching vehicle is acquired by subtracting the background screen of the measurement region prepared in advance from the input screen from the CCD camera 10. Spatial difference processing is performed to detect an edge portion (contour line) in which the brightness value changes abruptly in this image. The edge-detected image is binarized and delivered to the main processor 22 (steps S20 to S40 in FIG. 3).

The main processor 22 detects the presence and vehicle type of the vehicle by pattern-matching the contour image of the characteristic portion of the vehicle obtained by the binarization process with the reference contour image of the known vehicle type. . The position of the vehicle at the time of shooting is also calculated from the position of the image determined as the vehicle. These acquired data are stored in the local memory 23 together with the time obtained from the clock function of the main processor 22 (step S50 in FIG. 3). When the presence of the vehicle is confirmed (affirmative determination in step S60 of FIG. 3), the tracking process of the vehicle is executed by the main processor 22 (step S70 of FIG. 3).

In this tracking process, position data and speed data in the measurement area of the same vehicle are stored each time a photographing screen is input, and two vehicles come close to each other abnormally and one large vehicle is displayed on the screen. This is a process of predicting an abnormal situation that produces an image similar to the above, by using the stored history data. Details of this processing are shown in FIG. For the vehicle that has entered the measurement area for the first time, the local memory 23 does not store the history data regarding the position and speed. As a result, the main processor 22 determines that the vehicle has entered the measurement region for the first time (affirmative determination in step S200 of FIG. 4). The main processor 22 secures an area for storing the tracking history data of the vehicle in the local memory 23 for the initial processing, and stores the data indicating the vehicle position and the vehicle type obtained this time as the initial value of the history data. (Step S210 of FIG. 4).
In addition, to which position (referred to as a predicted position) the approaching vehicle will move at the time of the next image input is calculated with reference to the speed data of the history data of the preceding vehicle obtained so far. Since this calculation can be executed by a simple mathematical expression (current vehicle position + movement distance (speed × imaging interval time)), detailed description will not be required. A region in the screen corresponding to a region having a constant area centered on the obtained predicted position is a vehicle search region for the next time, in other words, a region in the screen for tracking the same vehicle (step S22 in FIG. 4).
0).

The vehicle that has entered the measurement area is captured by the CCD camera 10 multiple times while traveling in the measurement area. Second
The photographing screen obtained at the second time is preprocessed by the preprocessor 28 as described above, and delivered to the main processor 22 in the form of a binary image. The main processor 22 detects the vehicle position, speed, etc. in the set search area on the input screen (steps S20 to S50 in FIG. 3). The processing procedure proceeds from step S60 to step S70, and in the tracking processing of FIG. 4, the identification processing of the vehicle captured for the first time last time is performed. More specifically, a vehicle type determination process for confirming that the vehicle type of the vehicle detected this time matches the vehicle type of the vehicle detected previously is executed. Furthermore, the change in the speed of the vehicle before the present time is calculated, and it is confirmed that the speed of the vehicle this time has not changed more rapidly than the previous time. As an example, it is confirmed by comparison processing that the speed difference between the previous time and this time is smaller than the threshold value (steps S200 → S230, S2 in FIG. 4).
40). If the vehicle types do not match or if a rapid speed change is indicated, it is determined that the two vehicles are close to each other and that the vehicle was photographed as one vehicle in the photographing screen, and the abnormality handling after step S80 in FIG. 3 is performed. Move to processing.

On the other hand, when a normal determination is obtained in the above determination process, the detected position and speed ((current vehicle position-previous vehicle position) / acquired by calculating shooting interval time) are used as history data in the local memory. Add to 23. Further, the next search area is set (step S250 → S in FIG. 4).
260). Therefore, while the proximity state of the two vehicles does not occur and one vehicle is normally confirmed, the position and speed detected by the loop processing of steps S20 to S80 in the processing procedure of FIG. 3 are sequentially performed. Stored in the local memory 23. When the vehicle exits the measurement area, the exit is detected on the condition that the vehicle image disappears from the shooting screen and the previous vehicle position is located in the exit area. As a result, the tracking of the vehicle is terminated, and the conventional processing such as traffic flow calculation and parking / abnormal traffic detection is executed (steps S130 → S14 in FIG. 3).
0).

When an abnormality is detected in the tracking process of FIG. 4, the entire history data of the local memory 23 is searched to determine whether or not a following vehicle exists within a certain distance in the previous photographing screen. As an example, the distances between a plurality of vehicle positions stored in the local memory 23 and the vehicle currently being tracked and calculated are calculated, and it is determined whether or not the distances are smaller than a certain distance. If it is determined that there is no following vehicle, the process proceeds to the following tracking end determination processing (FIG. 4).
Step S90).

On the other hand, if it is determined that there is a following vehicle,
It is determined that two vehicles have been photographed as one vehicle, and the position and speed of the vehicle that has been determined to be abnormal are corrected by the following processing. In this embodiment, a vehicle traveling in the same direction cannot pass the preceding vehicle unless the lane is changed, and two vehicles are close to each other, and the inter-vehicle distance is substantially short during a very short period of 1 second or less. Focusing on the point that it is constant, when a situation occurs where the preceding vehicle cannot be tracked, the vehicle that cannot be tracked at this time is determined from the vehicle position of the following vehicle in the vicinity and the inter-vehicle distance determined from the two vehicle positions acquired last time. Calculate the position of (leading vehicle). For the speed, the correction value obtained by this calculation is used (the corrected vehicle position-
Calculated by calculating the (previous detected vehicle position) ÷ shooting interval time.
(Steps S100 → S110 in FIG. 4). In this way, even when two vehicles come close to each other and are photographed as one vehicle, it is possible to detect the two vehicles without performing a complicated image separation process as in the related art.

The following example can be implemented in addition to this embodiment.

1) In the present embodiment, the speed of the vehicle is calculated from the vehicle position acquired from the current screen, the vehicle position acquired from the previous screen, and the shooting interval time. May be used to calculate velocity.

2) In this embodiment, the image processing for one lane has been described, but when a plurality of lanes are included in the measurement area,
The same processing as described above is performed for each lane.

3) In this embodiment, when the vehicle is tracked, the area for searching the tracked vehicle is set only on the road portion in the input screen. As a result, one screen is not the search target, and high-speed processing can be realized. In order to improve the tracking accuracy, the size of the search area may be variably set in association with the vehicle speed and the change (acceleration) of the vehicle speed. For example, when the speed of the vehicle is high, the change in speed is large, so the search area is set, and when the speed of the vehicle is high, the search area is set large.

4) The vehicle positions at which two vehicles are photographed as one image differ depending on the installation position of the vehicle detector.
Further, depending on the orientation of the CCD camera, the preceding vehicle may be hidden by the following vehicle and the following vehicle may be hidden by the preceding vehicle. When the causal relationship between the hidden vehicle and the hidden vehicle is clear, the type of vehicle that should be corrected based on the positional relationship when there is only one vehicle (leading vehicle / subsequent vehicle)
May be determined by the main processor 22.

5) In order to detect an abnormal situation in which two vehicles are photographed as one vehicle, a method other than the method described in this embodiment can be used. For example, a method of determining that an abnormal situation occurs when a plurality of vehicle features are detected in the vehicle image on the input screen can be used.

[0032]

As described above, according to the first and fifth aspects of the present invention, an abnormal situation in which two vehicles are captured as one vehicle image is detected, and this one vehicle image is detected. Two
Considered as an image of one of the vehicles. Moreover, since the other vehicle image is calculated from the vehicle interval at the previous time point and the vehicle position at the current time point, the positions of the two vehicles can be detected by a simpler process than in the past. Also, the processing time is shortened.

In the second and third aspects of the present invention, a small vehicle is often hidden by a large vehicle following it when it is about to leave the measurement area. Therefore, one vehicle image is judged as a following vehicle, and the vehicle is hidden. The position of the vehicle is acquired using information obtained from the history data, for example, the inter-vehicle distance. As a result, the positions of two vehicles can be accurately grasped without separating the images of one vehicle.

According to the invention of claim 4, the speed is obtained from the vehicle position and the photographing interval time, and not only the vehicle position but also the speed of the adjacent vehicle is acquired. As a result, the vehicle detector can also be used as a speed measuring device.

[Brief description of drawings]

FIG. 1 is a perspective view showing a system installation state of an embodiment of the present invention.

FIG. 2 is a block diagram showing a system configuration of an embodiment of the present invention.

FIG. 3 is a flowchart showing a processing procedure of an embodiment of the present invention.

FIG. 4 is a flowchart showing details of vehicle tracking processing in step S70 of FIG.

FIG. 5 is an explanatory diagram for explaining a conventional vehicle detection method.

[Explanation of symbols]

 10 CCD Camera 11 Terminal Communication Unit 12 A / D Converter 13 I / O 20 Vehicle Detector Main Body 21 Image Processing Unit 22 Main Processor 23 Local Memory 24 CPU Bus 25 Dual Port Memory 26 Image Memory 27 Local Memory 28 Preprocessor 29 Image Bus

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication H04N 7/18 G06F 15/62 380

Claims (5)

[Claims] 1. A vehicle detector that photographs a vehicle that has entered a measurement region and measures the behavior of the vehicle from the image position of the vehicle image on the photographed screen shows the position of the same vehicle obtained in time series. A storage unit that stores history data for vehicle tracking, and an abnormality that detects an abnormal situation in which two vehicles traveling in the measurement area are close to each other and are photographed as one vehicle image in the photographing screen. When the abnormal situation is detected, the detecting means corrects the vehicle position of one of the two vehicles at the present time from the history data of the storage means before the current time when the abnormal situation was detected. And a processing unit, wherein one of the vehicle positions of the two vehicles is regarded as a corrected vehicle position and the other vehicle position is regarded as a vehicle position obtained from the one vehicle image, and the history at the present time is considered. Create data A vehicle detector characterized in that 2. The one vehicle position is a vehicle position of a preceding vehicle of the two vehicles, and the other vehicle position is a vehicle position of a succeeding vehicle of the two vehicles. The vehicle detector according to claim 1, characterized in that 3. The arithmetic processing means of claim 1 obtains an inter-vehicle distance of the two vehicles from the history data of claim 1 and the vehicle position of the following vehicle of claim 2, and the successor of claim 2 The vehicle position of the vehicle and the acquired inter-vehicle distance,
A vehicle detector for acquiring the vehicle position of the preceding vehicle. 4. The vehicle detector according to claim 1, wherein the arithmetic processing unit calculates the speed of the vehicle using the history data and the photographing interval time. 5. A vehicle detector for photographing a vehicle that has entered a measurement area and measuring the behavior of the vehicle from the image position of the vehicle image on the photographing screen. In the vehicle detection method when captured as a vehicle image, the position of the same vehicle obtained in time series from a plurality of photographing screens is stored in the vehicle detector as history data for vehicle tracking, and the measurement is performed. When an abnormal situation in which two vehicles traveling in the area approach each other and is photographed as one vehicle image in the photographing screen is detected, and when the abnormal situation is detected, the one vehicle image is The current vehicle position of the following vehicle is acquired by considering it as an image of the following vehicle of the two vehicles, the inter-vehicle distance of the two vehicles is acquired from the history data, and the current vehicle of the following vehicle is acquired. Position and the acquired car Vehicle sensing method for a vehicle detector, characterized in that the distance to acquire the vehicle position at the present time of the preceding vehicle.