阅读说明：本技术 一种检修工位上是否有车辆的检测方法 (Method for detecting whether vehicles exist on maintenance station ) 是由 陈强 沈修平 于 2020-04-23 设计创作，主要内容包括：本发明涉及一种检修工位上是否有车辆的检测方法,包括车辆物体一次检测以及基于车辆大小、车位大小的二次对比监测；所述检测方法采用的摄像头安装在监测车位的正前稍微偏上方位置,摄像头正对车位,进行定时抓图监测；所述检测方法首先对图像数据进行车辆物体监测,如果图像数据中无车辆物体,判断该车位无车辆；如果图像数据中有车辆物体,则进行二次的图片监测。通过对车辆位置的二次检测,可以提高车辆检修工位上车辆检测的准确度。(The invention relates to a method for detecting whether a vehicle exists on a maintenance station, which comprises the steps of carrying out primary detection on a vehicle object and carrying out secondary comparison monitoring based on the size of the vehicle and the size of a parking space; the camera adopted by the detection method is arranged at a position slightly above the front of a monitored parking space, and the camera is over against the parking space to perform timing image capturing monitoring; the detection method comprises the steps of firstly, carrying out vehicle object monitoring on image data, and judging that no vehicle exists in a parking space if no vehicle object exists in the image data; and if the image data contains the vehicle object, performing secondary picture monitoring. Through the secondary detection to the vehicle position, can improve the degree of accuracy that vehicle detected on the vehicle overhaul station.)

1.A detection method for detecting whether a vehicle exists on a maintenance station comprises primary detection of a vehicle object and secondary comparison monitoring based on the size of the vehicle and the size of a parking space; it is characterized in that the preparation method is characterized in that,

the camera adopted by the detection method is arranged at a position slightly above the front of a monitored parking space, and the camera is over against the parking space to perform timing image capturing monitoring;

the detection method comprises the steps of firstly, carrying out vehicle object monitoring on image data, and judging that no vehicle exists in a parking space if no vehicle object exists in the image data; if the image data contains the vehicle object, performing secondary picture monitoring;

the secondary picture monitoring process is as follows:

(1) the vehicle center point is inside the parking space position, and the vehicle width is smaller than the parking space position width; judging that the vehicle is in the parking space;

(2) the vehicle center point is inside the parking space position, and the vehicle width is greater than the parking space position width; judging that no vehicle is in the parking space; namely, the vehicle is suspected to drive through the parking space;

(3) the vehicle center point is below the parking space position, and the vehicle width is greater than the parking space position width; judging that no vehicle is in the parking space; namely, the vehicle is suspected to drive through the parking space;

(4) the vehicle center point is below the parking space position, and the vehicle width is less than the parking space position width; judging that the vehicle is in the parking space; namely, vehicles are suspected to be driven in or out from the parking spaces;

(5) the vehicle center point is above the parking space position, and the vehicle width is greater than the input maximum vehicle width; judging that the vehicle is in the parking space; i.e. the vehicle is suspended;

(6) the vehicle center point is above the interior of the parking space position, and the vehicle width is less than the input minimum vehicle body width; judging that no vehicle is in the parking space; the vehicle is suspected to be in a remote parking space (or in a remote parking space);

(7) if the vehicle center point is outside the parking space position, namely on the left or right, judging that the vehicle is not on the parking space;

the input data of the secondary picture monitoring comprises the following data:

the position area of the maintenance parking space on the picture is obtained through the size and the shooting position of the picture shot by the camera;

the maximum width and height and the minimum width and height of a vehicle displayed on a picture are obtained as input parameters when the vehicle is in a maintenance parking space through the size and the shooting position of a picture shot by a camera;

and monitoring the position of the vehicle, and calculating the position of the central point of the vehicle.

Technical Field

The invention relates to the field of AI image recognition, in particular to a method for recognizing whether a vehicle exists on a maintenance station.

Background

At present, AI image recognition technology wide application has appeared maintaining the parking stall in the middle of the production life in the vehicle maintenance trade and has whether has the detection demand of vehicle to help the enterprise carry out the management of maintaining the parking stall, and maintainer's management.

However, the conventional detection method is similar to a method for monitoring whether a vehicle is in a parking lot, and the method mainly depends on the following two points or a combination of the two points.

1. And detecting the license plate of the vehicle in the vehicle position area, detecting whether the license plate exists by using the traditional license plate detection technology, and judging whether the vehicle exists by judging whether the license plate exists. The disadvantage is that the meeting plate may be blocked during vehicle maintenance.

2. Vehicle object detects, directly carries out vehicle detection to the regional picture of parking stall, and more direct some.

However, the maintenance vehicle is different from the vehicle in the ordinary parking lot, and the suspended state of the vehicle often occurs in the maintenance process, which greatly affects the position judgment of the vehicle on the plane picture.

Object of the Invention

The invention aims to meet the requirement of detecting whether a vehicle is in a maintenance parking space in the vehicle maintenance industry.

The specific technical scheme of the invention is as follows:

a detection method for detecting whether a vehicle exists on a maintenance station comprises primary detection of a vehicle object and secondary comparison monitoring based on the size of the vehicle and the size of a parking space;

the camera adopted by the detection method is arranged at a position slightly above the front of a monitored parking space, and the camera is over against the parking space to perform timing image capturing monitoring;

the detection method comprises the steps of firstly, carrying out vehicle object monitoring on image data, and judging that no vehicle exists in a parking space if no vehicle object exists in the image data; if the image data contains the vehicle object, performing secondary picture monitoring;

the secondary picture monitoring process is as follows:

1. the vehicle center point is inside the parking space position, and the vehicle width is smaller than the parking space position width; judging that the vehicle is in the parking space;

2. the center point of the vehicle is positioned in the parking space position, and the width of the vehicle is greater than that of the parking space position; judging that no vehicle is in the parking space; namely, the vehicle is suspected to drive through the parking space;

3. the vehicle center point is below the parking space position, and the vehicle width is greater than the parking space position width; judging that no vehicle is in the parking space; namely, the vehicle is suspected to drive through the parking space;

4. the vehicle center point is below the parking space position, and the vehicle width is smaller than the parking space position width; judging that the vehicle is in the parking space; namely, vehicles are suspected to be driven in or out from the parking spaces;

5. the vehicle center point is above the parking position, and the vehicle width is greater than the input maximum vehicle width; judging that the vehicle is in the parking space; i.e. the vehicle is suspended;

6. the vehicle center point is positioned above the inner part of the parking space position, and the vehicle width is smaller than the input minimum vehicle body width; judging that no vehicle is in the parking space; it is suspected that the vehicle is in a remote parking space (or in a remote parking space).

7. And if the vehicle center point is outside the parking space position, namely on the left or right, judging that the vehicle is not in the parking space.

Further, the input data of the secondary image monitoring includes the following:

the position area of the maintenance parking space on the picture is obtained through the size and the shooting position of the picture shot by the camera;

the maximum width and height and the minimum width and height of a vehicle displayed on a picture are obtained as input parameters when the vehicle is in a maintenance parking space through the size and the shooting position of a picture shot by a camera;

and monitoring the position of the vehicle, and calculating the position of the central point of the vehicle.

Technical effects

In the vehicle maintenance trade, the management demand of having the maintenance parking stall appears, and it is long when the rate of utilization and the single maintenance that need real-time supervision out the maintenance parking stall, in order to prevent the maloperation of manual record and reduce the manpower, the AI monitoring mode that uses the camera to grab the picture in real time begins to appear, whether has the vehicle to maintain on the monitoring maintenance parking stall. Compared with the prior art, the invention has the technical effects that: through the secondary detection to the vehicle position, can improve the degree of accuracy that vehicle detected on the vehicle overhaul station.

Drawings

Fig. 1 is a schematic diagram of a method for detecting whether a vehicle is on a maintenance station according to the present invention.

Detailed Description

The following further describes the embodiments of the present invention with reference to the drawings:

and when a vehicle is in the parking space, the maximum width and height and the minimum width and height of the vehicle displayed on the picture are used as input parameters for detection.

And carrying out regular parking space photographing or object movement detection parking space photographing and sending the picture to the vehicle monitoring module.

Fig. 1 is a schematic diagram of a method for detecting whether a vehicle is present at a service station.

The vehicle monitoring module comprises the following two steps:

AI object detection: detecting a vehicle;

2. and (5) secondary detection of the vehicle position.

The detection method comprises the steps of firstly, carrying out vehicle object monitoring on image data, and judging that no vehicle exists in a parking space if no vehicle object exists in the image data; if the image data contains the vehicle object, performing secondary picture monitoring;

the secondary picture monitoring process is as follows:

1. the vehicle center point is inside the parking space position, and the vehicle width is smaller than the parking space position width; judging that the vehicle is in the parking space;

2. the center point of the vehicle is positioned in the parking space position, and the width of the vehicle is greater than that of the parking space position; judging that no vehicle is in the parking space; namely, the vehicle is suspected to drive through the parking space;

3. the vehicle center point is below the parking space position, and the vehicle width is greater than the parking space position width; judging that no vehicle is in the parking space; namely, the vehicle is suspected to drive through the parking space;

4. the vehicle center point is below the parking space position, and the vehicle width is smaller than the parking space position width; judging that the vehicle is in the parking space; namely, vehicles are suspected to be driven in or out from the parking spaces;

5. the vehicle center point is above the parking position, and the vehicle width is greater than the input maximum vehicle width; judging that the vehicle is in the parking space; i.e. the vehicle is suspended;

6. the vehicle center point is positioned above the inner part of the parking space position, and the vehicle width is smaller than the input minimum vehicle body width; judging that no vehicle is in the parking space; it is suspected that the vehicle is in a remote parking space (or in a remote parking space).

7. And if the vehicle center point is outside the parking space position, namely on the left or right, judging that the vehicle is not in the parking space.

In the aspect of vehicle detection of a maintenance parking space, the invention carries out secondary detection on the position of the vehicle on the basis of vehicle object identification, thereby improving the accuracy of vehicle detection.

The input data of the secondary picture monitoring comprises the following steps:

the position area of the maintenance parking space on the picture is obtained through the size and the shooting position of the picture shot by the camera;

the maximum width and height and the minimum width and height of a vehicle displayed on a picture are obtained as input parameters when the vehicle is in a maintenance parking space through the size and the shooting position of a picture shot by a camera;

and monitoring the position of the vehicle, and calculating the position of the central point of the vehicle.