JPS63205800A - Parking detection - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to a method for detecting the presence or absence of parking in a parking lot management system.

[Background] One important problem in configuring a parking lot management system is to know the parking situation in a parking lot as a positive value over time.

As mentioned above, if you can understand the parking situation, you can quickly instruct users where to park and
(If there are parking spaces available, we can collect the parking status of those parking lots, make judgments based on this information, and guide users to other appropriate parking lots.

Here, a conventional parking detection method of this type will be explained.

[Prior Art] Parking detection methods in parking lot management systems can be broadly classified into the following two methods as prior art.

(Method using vehicle detectors) This is a method of detecting parking by installing vehicle detectors one-on-one in parking areas.

(2) Method using the TV left camera This method installs the TV left camera in a position overlooking the parking area for multiple cars, processes the video signal of the TV left camera, and recognizes the presence of the vehicle itself. This is a method for detecting

[Problems to be Solved by the Invention] When using the method of Section 011 (1), vehicle detectors are installed one-to-one in the parking area. becomes. Therefore, not only does the equipment cost increase in proportion to the size of a large parking lot, but it also becomes more difficult to maintain and manage. In addition, especially when an ultrasonic vehicle detector is used, a large number of balls and arms are installed, which is undesirable from an aesthetic point of view.

1゛N To'1Document■0Method6　When doing It, TV#) 9One car 7・Multiple car parking 11 [Can be in charge of the area,
Moreover, if you switch the video signal and use it, the number of TV cameras will increase or decrease depending on the current school in the parking lot, but the image processing FI! It is possible to configure a system with a single processing device that performs several functions, and the problems raised in (2) will be resolved.

However, since it is necessary to detect stopped cars,
The method of detecting the presence of a vehicle from changes in brightness, which is normally used in the image processing type traffic flow meter 11 device, cannot be used. In other words, in an environment where there are changes in illuminance, it is necessary to process images on a static screen to recognize a wide variety of recognition objects in terms of species, size, color, shape, etc., and to recognize the existence of each object individually. No.

This is a factor that forces the detection algorithm to become complicated.

Ma5. In addition, under low illumination, the image from the vehicle may be
] Salmon becomes very indistinct, the SIN ratio becomes very poor, and not only does the detection accuracy deteriorate, but it may even become impossible to detect. For this reason, sufficient lighting equipment is required to prevent low illuminance even at night. This not only increases the equipment cost (not only does it cost a lot of money except when the 1st train enters and exits), but it also means that more lighting than necessary is given to people's places, which is a wasteful investment. do not have.

[Structure of the Invention] Of the three methods described above, the present invention
This was done for the purpose of solving the problems that conventionally occur or should occur in the f detection method, and the problem is solved by the following means.

First, a configuration example of a parking detection device using a TV camera in one parking lot will be explained with reference to FIG. In the figure, l is a TV camera, and 2 is a TV camera 1 that has multiple units.
A video signal switching device (MUX) for selecting one TV camera from among; 3 an A/D converter for digitalizing the video signal; 4 a memory board for storing image data; 5 performing image processing. Board 6 is a CPU board, 7 is a storage device for storing statistical information, etc., and 8 is an I10 board that controls input/output with the outside. ! The /6 board 8 has functions such as transmitting information to the -1- computer and controlling the information display board 9 used to provide information to users.

°"'In the parking lot management system described above, the parking detection method of the present invention allows 11
Rather than clearly marking a fixed-shaped mark and using a TV camera to detect the parked vehicle itself,
'Ki Mark's child;! This feature is characterized in that it isometrically determines whether or not the station is stationed by detecting 'rl+'.

The garrison detection method of the present invention will be explained in detail below with reference to the flowchart shown in FIG.

■Specify the TV camera to be processed first.

■Specify the parking area to be processed first.

■Determine whether or not the mark placed in the parking area is detected.

■If the mark of the parking area has been detected,
Indicates that there is nothing in the parking area that obstructs the visibility of the mark. In other words, it can be determined that there is no parking.

■The fact that the mark in the parking area cannot be detected indicates that something exists in the parking area that can prevent the mark from being visible. In other words, it can be determined that there is parking.

■Specify the parking area to be processed next and return to ■.

(2) Determine whether there are any remaining TV cameras to be processed. If there are none left, the process ends.

The TV camera to be processed in the 0th order is determined as th, and the process returns to step (2).

Note that the mark itself is provided with a light-emitting function, if necessary.

[Very diverse in terms of vehicle type, size, color, shape, etc.
[Instead of recognizing both objects themselves, the present invention recognizes marks whose positions, sizes, colors, and shapes are specific and known, so the detection algorithm is more sophisticated than in the past.

If the mark itself has a light-emitting function, it will be low! ! By emitting light from the mark itself even at a temperature below 6 degrees, the image produced by the mark can maintain its clarity.

An example of a parking lot with parking areas for only 1 and 5 cars will be described.

FIG. 3(a) shows a state in which there are no vehicles in parking areas 21 to 25 of the parking lot, and FIG.
+ and 22.25 respectively.

A TV camera 1 and an illuminance sensor IO are attached near the top of the pole 12. Three marks to 35 are provided in each parking area 21 to 25, corresponding to the parking areas 21 to 25. Sukuna (marks 31-35 are TV camera 1)
The TV camera 1 is positioned so that it is within the field of view.

In FIG. 3(a), since there is no vehicle in the iIL parking lot and all marks 31 to 35 are visible by the TV camera 1, the number of parked vehicles is determined to be zero.

In FIG. 3(b), only two marks 33 and 34 are visible by the TV camera 1, so the number of parked cars is determined to be three.

In detecting a mark, it is necessary that the mark be clear. Therefore, in parking lots where there is insufficient lighting such as at night, visibility can be improved even under low illumination by providing the mark itself with a light-emitting function.

As shown in FIG. 4, whether or not to light up the mark is determined depending on the level of illuminance. In other words, when the illuminance drops below a constant, the mark is lit.

In order to realize this function, as shown in FIG. As mentioned above, an example of the flow of processing near the top of the pole 12 in FIG. 3 will be explained with reference to the flowchart in FIG. 6 with reference to FIG. 2.

■Image data is imported into the memory on the TV TV camera 1 memory board 3.

■ Parking area 2I is the first processing area.

(2) Binarize the image data in the processing area using the calculation board 5.

(2) The degree of matching between the pre-stored mark pattern of the processing area and the /dimension turn obtained by binarization is calculated by the calculation board 5.

■If the degree of matching is greater than a predetermined constant, it is determined that there is a mark, otherwise it is determined that there is no mark.

■When the mark is detected, it is determined that there is an unparked vehicle.

- Determine whether parking area 25 has been completed.

Set the 0th order processing area.

■Marks are not detected even for vehicles moving in a parking area, so only parking areas that have been determined to be parked twice in a row within a certain period of time are considered to have been confirmed to be parked. Then, the information display board 9 displays that the vehicle is parked at 11L.

[Effects of the Invention] As explained above, the present invention solves the conventional problems while maintaining the superiority of the detection method using a TV camera over the conventional detection method using a vehicle detector. I was able to resolve the issue. In other words, the present invention recognizes marks whose positions, sizes, colors, and shapes are specific and known, rather than recognizing vehicles themselves, which vary widely in terms of vehicle type, size, color, and shape. Since the detection algorithm is not only easier than the conventional method, but also the detection ti' is improved compared to the conventional method using a TV camera.

Furthermore, the simplification of the detection algorithm simplifies the hardware circuit of the arithmetic board 5 that performs image processing, thereby maintaining sufficient clarity of the image.

In this case, special rules 1 for parking management system applications are required.
Stable flame emission accuracy can be obtained all day long without the need for a brightening device or the like.

[Brief explanation of the drawing]

FIG. 1 is a flowchart for explaining the parking detection method of the present invention. Figure 2 shows a TV in one parking lot to which the present invention is applied.
This is an example of a configuration of a parking detection device using a camera. FIG. 3 shows an embodiment of the present invention, in which (a) shows a case where there is no parking, and (b) shows a case where there is no parking. FIG. 4 is a flowchart showing mark lighting processing. FIG. 5 is an explanatory diagram of the rlo board. FIG. 6 is a flowchart for explaining the present invention in detail. ■...TV camera, 2...Video signal switching circuit, 3.
...A/D converter, 4...Memory board, 5...
・Calculation board, 6...CPU board, 7...Storage device, 8...Ilo board, 9...Information display board, 1G...Illuminance sensor,! ! ...Mark light fixture, 1
2...Paul. 91st r-geki% 3 Figure (α) 14 Figure 5 Figure 1υ II

Claims (2)

[Claims] (1) A clear mark is placed in each parking area of each car in the parking lot, and a TV camera detects the mark in the parking area.If the mark in the parking area is not detected, the parking lot is A parking detection method characterized by determining that the vehicle is parked in an area. (2) The parking detection method according to claim 1, characterized in that marks each having a light emitting function are provided within the parking area.