JPH1097697A - Method and device for detecting parking vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To exactly judge a parking vehicle with a simplified structure by extracting an image of a graphic by drawing the graphic on a parking position of a parking lot and processing an image on which the parking lot is picked up. SOLUTION: A marker 6 for recognizing vehicle is formed in a prescribed graphic arranged in a parking space 5. Presence/absence of the image of the marker 6 for recognizing vehicle is discriminated by processing the image on which the parking lot is picked up by a photographing device 7 and the presence/absence of the vehicle in the parking space 5 is outputted by an image processor 8. When the parking vehicle exists on the parking space 5, the marker 6 for recognizing vehicle is shielded and in the result of the image processing, the picture of the graphic corresponding to the marker 6 for recognizing vehicle is not extracted. Consequently, the presence/absence of the parking vehicle on the parking space 5 is easily discriminated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for detecting both vehicles which are parked at a parking place in a parking lot, and more particularly to a method for detecting both vehicles which can be accurately determined with a simple structure. The present invention relates to a method and an apparatus.

[0002]

2. Description of the Related Art As a method of automatically detecting a parked vehicle (hereinafter referred to as both parking vehicles), for example, application of a vehicle detector using radiation waves such as microwaves and ultrasonic waves can be considered. In this method, for example, as shown in FIG. 3, a radio wave transmitted from a radio wave transmitter / receiver 1 installed at a lower part or a side wall of a parking lot is radiated in a direction that both parkings should be, reflected by both parkings, and transmitted to the radio wave transmitter / receiver 1 Detect the returning radio wave. If there is both parking spaces, the presence of both parking spaces is detected because the received signal strength is high. When there is no parking both, since the received signal strength is small, it is detected that there is no parking both. For this purpose, a threshold value of the received signal strength is set, and the presence / absence of both parking spaces is determined by comparing the threshold value with the threshold value.

[0003] As a similar method to the above method, there is an optical system using light instead of radio waves. In this case, a transmitter / receiver is used instead of the radio wave transmitter / receiver 1, and the received signal strength is used as the received signal strength. The method of determining the presence or absence of both parking spaces is almost the same as the above method.

On the other hand, there is another method using image processing. This method focuses on two consecutive images of an image obtained by continuously photographing a parking place (hereinafter, referred to as a parking space), and extracts a size, a color, and the like of the vehicle, which are characteristic amounts of the vehicle, from a difference between the two images. It is determined whether or not the image extract obtained by such image processing is a vehicle, and recognizes both parking spaces.

[0005]

However, these methods have the following drawbacks.

First, in the method using a vehicle detector such as a radio transmitter / receiver or a light transmitter / receiver, it is necessary to install a large number of vehicle detectors along a parking place in order to detect the situation of the entire parking lot. In addition, a transmission device 2 for transmitting data from these many vehicle detectors to one place and a transmission path 3 such as an electric wire or an optical cable are required. Then, a display / output device 4 and software for displaying / outputting the full / vacant information of the entire parking lot using the received data are required, and the overall cost increases.

On the other hand, the method using image processing has the advantage that a single camera can detect a large number of parking lots, but on the other hand, accurately determines whether or not the image extract obtained by the image processing is a vehicle. Has not been established yet, and it has not been put to practical use because there are many misrecognitions / non-recognitions. For this problem, several ideas have been proposed. For example, as a countermeasure against erroneous recognition, there is a method of judging based on a difference in the size of the image extractor. However, in this method, it may be difficult to distinguish a family group or the like that relatively moves together and moves from a small passenger car.

[0008] Further, there is a method of identifying the vehicle body color based on the vehicle body color. However, when it is difficult to distinguish between the color of the asphalt at the parking place and the color of the black vehicle body, or when the color detected by the screen luminance is different. In some cases, it may not be possible to determine the presence or absence of a vehicle depending on conditions. In any case, a practical system cannot be realized by this method.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above-mentioned problems and to provide a method and an apparatus for detecting both parking spaces, which can make a precise determination with a simple configuration.

[0010]

In order to achieve the above object, according to the present invention, a predetermined figure is written at each parking place of a parking lot,
An image of the figure is extracted by performing image processing on an image of the parking lot, and the presence or absence of the parked vehicle is detected based on the presence or absence of the image.

The figure may have a deformation and a change in size according to the positional relationship between the parking place where the figure is described and the place where the image is taken.

The apparatus includes a marker arranged at each parking place in a parking lot and formed in a predetermined figure, a photographing apparatus for photographing the parking lot, and image processing of an image photographed by the photographing apparatus. And an image processing device for extracting an image of the marker and detecting the presence or absence of a parked vehicle based on the presence or absence of the image.

[0013]

An embodiment of the present invention will be described below in detail with reference to the accompanying drawings.

As shown in FIG. 1, a parking both detecting device according to the present invention includes a vehicle recognition marker 6 arranged in each parking space 5 and formed in a predetermined figure, an ITV camera for photographing a parking lot, and the like. And an image processing device that performs image processing on an image photographed by the photographing device 7 to identify the presence or absence of the image of the vehicle recognition marker 6, and finally outputs the presence or absence of the vehicle in each parking space 5. And 8. When the distance between the imaging device 7 and the image processing device 8 is large, an image transmission device may be inserted between them.

The vehicle recognition marker 6 is marked on the road surface in the parking space 5 by applying paint of a predetermined color or by bonding a baked resin plate used for specifying road signs and pedestrian crossings. You. Further, in the present invention, the deformation and the change in the size of the vehicle recognition marker 6 for each parking space 5 are determined based on the positional relationship between the parking space 5 and the photographing device 7.

As shown in FIG. 2, the photographing device 7 is arranged at a position overlooking the road surface so that the vehicle recognition marker 6 is included in the image, and can photograph a plurality of parking spaces 5.

For the sake of simplicity, it is assumed that the shape of the vehicle recognition marker 6 on the image of the photographing device 7, that is, the image of the vehicle recognition marker 6 is circular, and the two parking spaces 5 Consider the shape of the vehicle recognition marker 6 arranged in the. In FIG. 2, the marker images S1 and S2 projected at the same distance from the center of the photographing device 7 are assumed to be circular with the same size on the image. These marker images S
1 and S2 are extended at the angle of view so that one end is in contact with the horizontal axis indicating the road surface. The marker images S1 ', S at this time
Actual vehicle recognition marker 6a and 2 ', ₁ an angle between 6b theta, When theta _2, theta _1, a vehicle recognition marker 6a in accordance with the theta _2, 6b becomes flattened and elliptical than circular. Coefficients S6a and S6b indicating deformation and size change of the vehicle recognition markers 6a and 6b are calculated as follows, respectively.

S6a = S1 ′ / sin θ ₁ S6b = S2 ′ / sin θ ₂ Here, the molecules S1 ′ and S2 ′ are geometrically obtained from the ratio of the distance between the parking space 5 and the photographing device 7. .

As can be seen from the above equation, the shape of the actual vehicle recognition markers 6a, 6b is deformed due to the difference in the angle of the photographing device 7 with respect to the parking space 5, and the parking space 5
As the distance between the vehicle and the photographing device 7 increases, the actual shape of the vehicle recognition markers 6a and 6b increases. In this way, the deformation and the change in the size of the vehicle recognition marker 6 for each parking space 5 are determined by the parking space 5 and the photographing device 7.
Is determined by the positional relationship with

Although the description has been made assuming that the image of the vehicle recognition marker 6 is circular, the shape is not particularly limited to a circular shape, and any shape can be selected.

Next, the processing contents of the image processing device 8 will be described, and the reason why the image of the vehicle recognition marker 6 is configured to be the same for a plurality of parking spaces 5 will be described.

The parking space 5 in which the vehicle recognition marker 6 is installed is photographed by the photographing device 7 and the image is taken by the image processing device 8.
To enter. In the image processing device 8, the input image is processed by a series of image processing methods (for example, input → division into each pixel → smoothing of the image → binarization → labeling → area extraction for each label → operation / judgment). The presence or absence of the vehicle recognition marker 6 is output. That is, in the image processing method in the both-parking detection device of the present invention, an image of a figure corresponding to the vehicle recognition marker 6 is obtained from an image of the vehicle recognition marker 6 having a different size / deformation shape for each parking space 5. Extract. From the presence or absence of this image, the presence or absence of both parking spaces on the parking space 5 is detected.

As a matter of course, if there is a parking space in the parking space 5, the vehicle recognition marker 6 is shielded, so that the vehicle recognition marker 6 is not reflected on the image photographed by the photographing device 7, and the image processing result The image of the figure corresponding to the vehicle recognition marker 6 cannot be extracted. Conversely, if there are no parking spaces on the parking space 5, an image of a predetermined figure is extracted. Therefore, the presence or absence of both parking spaces on the parking space 5 can be easily identified. By collecting the information on the presence or absence of both parking spaces in each of the parking spaces 5 in one place, and detecting the availability of all the parking spaces 5 and the absence (full) of each of the parking spaces 5, it is possible to detect the fullness of the entire parking lot. .

In the image processing, if the extracted figure becomes complicated in size and shape, a complicated method must be used to determine whether the figure corresponds to the vehicle recognition marker 6 or not. . Such a complicated method requires many processing procedures and a long calculation time. In this respect, according to the present invention, the vehicle recognition marker 6 having various deformations and size changes installed in each parking space 5 is photographed by the photographing device 7 and is input to the image processing device 8. The shape and the size are almost the same. Thereby, it is significantly easier to determine whether the vehicle recognition marker 6 is included in the image. As a result, the processing procedure can be simplified and the calculation time can be reduced.

According to the present invention, it is only necessary to recognize the vehicle recognition marker 6 whose location and shape are determined in advance, so that the processing is performed in comparison with the conventional method of recognizing a vehicle having various shapes and colors. It will be easier. In addition, by making the shape of the vehicle recognition marker 6 captured by the image capturing device 7 on the image constant regardless of the position of the parking space 5, the image processing device 8 may extract an image of the fixed shape. Therefore, the processing content of the image processing device 8 is further greatly simplified.

[0026]

The present invention exhibits the following excellent effects.

(1) It is not necessary to separately install vehicle detectors, so that the system is simple and the image processing apparatus is greatly simplified, so that the cost is reduced.

(2) Since it is sufficient to recognize a fixed figure,
You can make accurate decisions.

[Brief description of the drawings]

FIG. 1 is a perspective view of a parking both detecting device according to an embodiment of the present invention.

FIG. 2 is a side view of the parking both detecting device of FIG. 1;

FIG. 3 is a plan view of a conventional parking both detecting device.

[Explanation of symbols]

 5 Parking space (parking place) 6 Marker for vehicle recognition 7 Imaging device 8 Image processing device

Claims (3)

[Claims] 1. A predetermined graphic is written at each parking place of a parking lot, and an image of the graphic is extracted by processing an image of the parking lot, and an image of the graphic is extracted. A method for detecting both parkings, characterized by detecting the presence or absence. 2. The parking apparatus according to claim 1, wherein the figure has a deformation and a change in size according to a positional relationship between a parking place where the figure is written and a place where the image is taken. Both detection methods. 3. A marker arranged at each parking place of a parking lot and formed in a predetermined figure, a photographing device for photographing the parking lot, and an image photographed by the photographing device for image processing of the marker. An image processing device that extracts an image and detects the presence or absence of a parked vehicle based on the presence or absence of the image.