JP2000137896A - Parking lot jam situation judgment device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To more precisely and automatically judge a jam situation in a parking lot such as a rest house of a freeway with few cameras. SOLUTION: A parking lot is photographed by plural ITV cameras 1. Pictures photographed by the ITV cameras 1 are individually analyzed by a picture processing part 2 and the number of parking vehicles is obtained. A parking rate is obtained at every ITV camera from the number of parking vehicles and the number of parking squares. A parking rate estimation part 3 weights/averages the parking rate at every ITV camera in accordance with a photographing object area and the whole estimation parking rate is obtained. A parking rate correction part 4 corrects the estimation parking rate by a correction function corresponding to the parking lot and a corrected parking rate is obtained. The estimated parking rate is corrected with the correction function with which the corrected parking rate becomes 80% when the estimated parking rate is 80%, and the corrected parking rate becomes 20% when the estimated parking rate is 0%. A jam degree judgment part 5 compares the corrected parking rate with a threshold and obtains a jam degree. The parking situation of the parking lot except for a monitored area is estimated and the jam situation is corrected. The jam situation can be inexpensively judged. The lowering of the parking rate due to the bias of the parking vehicles in the parking lot is compensated and the corrected parking rate approximated to a real state can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a parking lot congestion status judging device, and more particularly to a parking lot congestion status judging device capable of accurately estimating a congestion rate.

[0002]

2. Description of the Related Art The number of users of highway parking lots increases with increasing traffic, and parking lots are becoming more crowded. In order to alleviate the concentration in a specific parking lot and to level the usage and improve efficiency, parking lot congestion information is provided to vehicles traveling upstream of the parking lot.

In the initial congestion status information collection method, a vehicle detector is installed at an entrance / exit ramp of a parking lot, and the number of vehicles in the parking lot is determined from the difference in the number of vehicles measured at the entrance / exit to determine the parking situation. Was used. However, measurement errors due to vehicle detectors accumulate over time,
There is a problem that correction by a person is required. Recently, multiple ITV cameras have been installed in parking areas,
A method of calculating the number of parked vehicles by performing image processing on an image captured by a V camera is used.

As a method of monitoring a parking area, there are a method in which the entire area is monitored, and a method in which some representative areas are monitored. In the former, the entire area is monitored, so that the determination accuracy is high, but the number of cameras and the image processing capacity increase, and the equipment cost increases. In the latter case, the determination accuracy is reduced, but the equipment cost is reduced. However, "full"
The latter method is often used because high judgment accuracy is not required, such as "congestion" and "empty".

Hereinafter, a conventional congestion status information collecting apparatus by image processing will be described with reference to FIG. In FIG. 5, an ITV camera 1 is a camera that photographs a parking area in a predetermined range. The image processing unit 2 is a means for processing the image of the ITV camera 1 to obtain the number of vehicles. The parking rate estimating unit 3 is means for estimating the parking rate of the entire parking lot from the parking rate of each ITV camera. The congestion degree determination part 5 is a means for comparing the estimated parking rate with a threshold to determine the congestion degree.

[0006] ITV camera 1 installed in each parking area
Is transmitted to the image processing unit 2. The image processing unit 2 determines the presence or absence of a vehicle for each parking space and outputs the number of parked vehicles. For example, for a digital image, the edge information of the image information is emphasized using a differentiation method, and a feature amount space threshold value set in advance based on the average brightness value, the variance value, and the differentiation value in the parking space is used. Is used to determine the presence or absence of a vehicle in the parking space.

The parking rate estimating unit 3 calculates the parking rate of the parking space for each camera and the parking rate of the entire parking lot from the information on the number of parked vehicles determined by the image processing unit 2. The parking rate for each camera is (number of vehicles recognized / number of target parking squares). The parking rate of the parking lot is a value obtained by multiplying the parking rate of each camera by the correction rate of each camera, obtaining the sum, and dividing the sum by the number of cameras.

[0008] The congestion degree determination unit 5 determines the congestion degree as "full".
The result of the determination in three ranks of "congestion" and "empty" is transmitted to the central unit. When the parking rate is 0 to 70%, it is determined to be “empty”, and when it is 70 to 95%, it is determined to be “congested”.
When it is 100%, it is determined that the vehicle is full. The judgment results are collected in the central device, and the “full” /
Display "congestion" / "empty".

[0009]

However, in the conventional parking lot congestion state judging device, the parking ratio measured value for each camera is weighted and averaged to obtain the entire parking ratio. There is a problem that an error due to the deviation is large and it is determined that there is congestion but it is empty, or that it is determined that there is congestion even though it is not congested. In addition, in order to obtain a reliable parking rate, it is necessary to photograph a wide range, and there is a problem that the number of cameras to be installed increases and costs increase.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above-mentioned conventional problems and to automatically and accurately determine the congestion state in a parking lot with a small number of cameras.

[0011]

In order to solve the above-mentioned problems, according to the present invention, a parking lot congestion state judging device is provided with a plurality of ITV cameras for photographing a parking lot and an image taken by each ITV camera individually. An image processing apparatus for calculating the number of parked vehicles by analyzing the number of parked vehicles, and a calculating means for calculating a parking ratio for each ITV camera from the number of parked vehicles and the number of parking spaces;
Estimating means for obtaining an overall estimated parking rate by weighting and averaging the parking rate of each TV camera in accordance with the shooting target area, and correcting for obtaining the corrected parking rate by correcting the estimated parking rate with a correction function corresponding to the parking lot Means and a congestion degree determination device that obtains the congestion degree by comparing the corrected parking rate with a threshold value.

[0012] With this configuration, it is possible to estimate the parking situation of the parking lot outside the monitoring area, perform the congestion situation correction processing, and determine the congestion situation at low cost.

[0013] The correction means is provided with means for correcting with a correction function such that the corrected parking rate becomes 80% when the estimated parking rate is 80%, and the corrected parking rate becomes 20% when the estimated parking rate is 0%. And

[0014] With this configuration, it is possible to obtain a corrected parking rate close to the actual situation by compensating the calculation accuracy of the parking rate due to the bias of the parked vehicle.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION
A plurality of ITV cameras for photographing a parking lot, an image processing device for individually analyzing images photographed by each ITV camera to obtain the number of parked vehicles, and an ITV based on the number of parked vehicles and the number of parking spaces. Calculating means for obtaining a parking rate for each camera; estimating means for obtaining an overall estimated parking rate by weighting and averaging the parking rate for each ITV camera in accordance with a shooting target area; and correcting the estimated parking rate according to the parking lot A parking lot congestion status determination device comprising a correction means for obtaining a corrected parking ratio by correcting with a function, and a congestion degree determination device for obtaining a degree of congestion by comparing the corrected parking ratio with a threshold value. Has the effect of estimating the parking rate and correcting it according to the characteristics of the parking lot to obtain a more accurate parking rate.

The second aspect of the present invention is the first aspect of the present invention.
In the parking lot congestion situation determination device described in the above, the correction means, when the estimated parking rate is 80%, the corrected parking rate is 80%,
When the estimated parking rate is 0%, the corrected parking rate is corrected by a correction function that is 20%. The corrected parking rate is compensated for a reduction in the parking rate obtained from an image taken by a camera, and a more accurate correction is performed. It has the effect of finding the parking rate.

According to a third aspect of the present invention, a plurality of I
The parking lot is photographed with a TV camera, and the images photographed by each ITV camera are individually analyzed by an image processing device to obtain the number of parked vehicles, and I is calculated from the number of parked vehicles and the number of parking spaces.
The parking rate is determined for each TV camera, the parking rate for each ITV camera is weighted and averaged according to the shooting target area to determine the overall estimated parking rate, and the estimated parking rate is corrected by a correction function corresponding to the parking lot. It is a parking lot congestion situation determination method of obtaining a corrected parking rate, comparing the corrected parking rate with a threshold to obtain a congestion degree,
This has the effect of estimating the parking rate from images captured by a small number of cameras, correcting the parking rate according to the characteristics of the parking lot, and obtaining a more accurate parking rate.

FIG. 1 shows an embodiment of the present invention.
This will be described in detail with reference to FIGS.

(Embodiment) In the embodiment of the present invention, the parking lot is photographed by a plurality of ITV cameras, and the images photographed by the respective ITV cameras are individually analyzed by an image processing unit, and the number of parked vehicles is determined. Is obtained, the parking rate is determined for each ITV camera from the number of parked vehicles and the number of parking squares, and the parking rate for each ITV camera is weighted and averaged according to the imaging target area to obtain an overall estimated parking rate. When the estimated parking rate is 80%, the corrected parking rate is 80%, and when the estimated parking rate is 0%, the corrected parking rate is 20%. This is a parking lot congestion state determination device that obtains a congestion degree by comparing the calculated parking rate with a threshold value.

FIG. 1 is a functional block diagram of a parking lot congestion state judging device according to an embodiment of the present invention. In FIG.
The ITV camera 1 is a camera that photographs a parking area in a predetermined range. The image processing unit 2 is a device that processes an image of the ITV camera 1 and obtains the number of vehicles. Parking rate estimation unit 3
Is means for estimating the parking rate of the entire parking lot from the parking rate of each ITV camera. The parking ratio correction unit 4 is a unit that corrects the estimated parking ratio with a correction function to obtain a corrected parking ratio. The congestion degree determination unit 5 is a means for comparing the corrected parking rate with a threshold to determine the congestion degree. FIG. 2 is a processing flowchart of the parking lot congestion state determination device according to the embodiment of the present invention.

The operation of the parking lot congestion state judging device according to the embodiment of the present invention configured as described above will be described. A plurality of ITV cameras 1 for photographing several parking areas of a parking lot are installed. The images of the ITV camera 1 installed in each parking area are sequentially switched and transmitted to the image processing unit 2.

The image processing unit 2 determines the presence or absence of a vehicle for each parking space and calculates the number of parked vehicles for each camera. Image processing is performed in the same manner as in the prior art to identify the vehicle and determine the number of parked vehicles.

In S1 (step 1) of FIG. 2, the parking rate estimating section 3 calculates the parking rate for the parking space for each camera from the information on the number of parked vehicles determined by the image processing section 2. In S2, the estimated parking rate of the entire parking lot is calculated. The parking rate of the entire parking lot is a value obtained by multiplying the parking rate of each camera by a weighting factor of each camera, obtaining the sum, and dividing the sum by the number of cameras. This is set as the estimated parking rate.

As shown in FIG. 3, the characteristics of the parking ratio differ depending on the shooting target area for each camera. In a convenient parking area near the cafeteria, the parking space is filled at the beginning when the overall parking rate is low. Therefore, even if the overall parking rate is low as in the curve A, the measured value of the parking rate is large. Conversely, in an inconvenient parking area away from the cafeteria and toilet, the parking space is filled at the end of the whole, so the measured value of the parking rate remains small until the overall parking rate increases, as shown by curve C. is there. At the middle point, the area is filled almost corresponding to the entire parking rate, and the curve B is obtained. It is good to use only measured value data that is close to the overall parking rate. However, because the number of samples is small, where the parking rate measured by the ITV camera matches the overall parking rate, the weight is increased and added. By adding small weights with small weights, it is possible to obtain an estimated value with a small error as a whole.

In S3, the parking rate correction unit 4 calculates a corrected parking rate from the estimated parking rate. When the estimated parking rate is 80%, the corrected parking rate is 80%, when the estimated parking rate is 0%, the corrected parking rate is 20%, and when the estimated parking rate is 100%, the corrected parking rate is 95%. to correct.

In the measurement with the ITV camera, the measured value of the parking ratio becomes lower than the actual value due to the deviation of the parked vehicle in the parking lot. A coefficient for correcting the measured parking rate with the ITV camera is determined by comparing the measured parking rate with the ITV camera and the result of actually measuring the parking rate manually. If a fixed value is added uniformly and corrected, the approximation will not always be a good approximation, so if the parking rate is small, the correction will be large, and if the parking rate is large, the primary will be almost the same so that they match at 80%. Correct by approximation. FIG. 4 shows a graph of the correction function. Even if a complicated correction function having a quadratic function or more is used, only the calculation becomes complicated, and the exceptional accuracy is not improved.

The congestion degree determination unit 5 sets the congestion degree to “full”.
It is divided into three ranks: "congested" and "empty". Parking rate 95 ~
When it is 100%, it is determined that the vehicle is full in S4. When the parking rate is 70 to 95%, it is determined that the traffic is "congested" in S5. When the parking rate is 0 to 70%, it is determined in step S5 that the vehicle is empty. At S6, the congestion degree is output and displayed.

As described above, in the first embodiment of the present invention, the parking lot congestion state judging device uses the plurality of ITV cameras to photograph the parking lot and individually processes the images photographed by each ITV camera. Department to analyze and find the number of parked vehicles,
The parking rate is determined for each ITV camera from the number of parked vehicles and the number of parking squares, and the parking rate for each ITV camera is weighted and averaged according to the shooting target area to obtain the overall estimated parking rate. Is 80%, the corrected parking rate is 80%, and when the estimated parking rate is 0%, the corrected parking rate is 20%. The parking ratio is compared with the threshold to determine the degree of congestion, so the parking situation of the parking lot outside the monitoring area is estimated and the congestion situation is corrected to obtain a corrected parking rate that is close to the actual situation, and low cost is obtained. To determine the congestion status.

[0029]

As described above, according to the present invention, a parking lot congestion state judging device is provided with a plurality of ITV cameras for photographing a parking lot, and an image taken by each ITV camera is individually analyzed for a parked vehicle. Image processing means for calculating the number of vehicles, computing means for calculating the parking rate for each ITV camera from the number of parked vehicles and the number of parking squares, and weighting and averaging the parking rate for each ITV camera in accordance with the shooting target area Means for obtaining an estimated parking rate of the entire vehicle, correction means for correcting the estimated parking rate by a correction function corresponding to the parking lot to obtain a corrected parking rate, and congestion for comparing the corrected parking rate with a threshold to obtain a congestion degree. With the configuration including the degree determination means, the congestion state can be accurately determined even with the minimum camera arrangement, and the effect that the congestion state can be determined at a lower cost than before by reducing the number of cameras. Get That.

Further, the correction means is provided with means for correcting with a correction function such that when the estimated parking rate is 80%, the corrected parking rate is 80%, and when the estimated parking rate is 0%, the corrected parking rate is 20%. Therefore, it is possible to obtain the effect of compensating the measured value based on the camera image and determining the congestion state at a low cost with the corrected parking rate close to the actual state.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a parking lot congestion state determination device according to an embodiment of the present invention;

FIG. 2 is a processing flowchart of the parking lot congestion state determination device according to the embodiment of the present invention;

FIG. 3 is a graph showing the relationship between the measured parking rate for each camera and the overall parking rate;

FIG. 4 is a graph of a conversion formula for obtaining a corrected parking rate from an estimated parking rate,

FIG. 5 is a functional block diagram of a conventional parking lot congestion state determination device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ITV camera 2 Image processing part 3 Parking rate estimation part 4 Parking rate correction part 5 Congestion degree determination part

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Mitsuaki Ota 4-3-1 Tsunashima Higashi, Kohoku-ku, Yokohama, Kanagawa Prefecture Inside Matsushita Communication Industrial Co., Ltd. No.3-1, Matsushita Communication Industrial Co., Ltd. F-term (reference) 5H180 AA01 CC04 KK01 KK03 KK06 KK10

Claims (3)

[Claims] A plurality of ITV cameras for photographing a parking lot; image processing means for individually analyzing images photographed by the respective ITV cameras to obtain the number of parked vehicles; Calculating means for obtaining a parking rate for each ITV camera from the number of cells; estimating means for obtaining a total estimated parking rate by weighting and averaging the parking rates for each ITV camera in accordance with an imaging target area; A parking lot congestion state determination comprising: a correction unit that obtains a correction parking ratio by correcting with a correction function corresponding to a parking lot; and a congestion degree determination device that obtains a degree of congestion by comparing the correction parking ratio with a threshold. apparatus. 2. The correction means includes means for correcting with a correction function such that when the estimated parking rate is 80%, the corrected parking rate is 80%, and when the estimated parking rate is 0%, the corrected parking rate is 20%. The parking lot congestion state determination device according to claim 1, wherein: 3. The parking lot is photographed with a plurality of ITV cameras,
The images taken by each ITV camera are individually analyzed by the image processing means to obtain the number of parked vehicles, the parking rate is obtained for each ITV camera from the number of parked vehicles and the number of parking spaces, and each ITV camera is obtained. The estimated parking rate is calculated by weighting and averaging the parking rates according to the shooting target area, and the estimated parking rate is a correction function according to the parking lot. When the estimated parking rate is 80%, the corrected parking rate is 80% A parking lot congestion state determination method characterized in that when the estimated parking rate is 0%, the corrected parking rate is corrected to be 20%, and the corrected parking rate is compared with a threshold to determine the congestion degree.