Gate based on combination of camera and speed measurement sensor

Abstract

The utility model discloses a gate based on combination of a camera and a speed measurement sensor, which comprises two stand columns arranged at intervals, a gate passage formed between the two stand columns, a gate capable of being opened and closed mounted in the gate passage, a gate connection control system, and a camera and a speed measurement sensor mounted above the gate passage to form a three-dimensional image of an image and a point cloud. The control system comprises a calibration module, a detection tracking module, a judgment module and a gate control module, the calibration module is used for carrying out alignment calibration on the camera and the speed measurement sensor, and the detection tracking module is connected with the camera and used for detecting and tracking a head or a human body in a video or an image acquired by the camera; the judgment module is connected with the detection tracking module and the speed measurement sensor and used for judging whether the detection area is a human body or other objects; the gate control module is connected with the gate to control opening and closing of the gate. According to the utility model, the number of passing people can be accurately calculated, trailing false alarms are reduced or even eliminated, and the passing efficiency and safety of the gate are improved.

Description

Technical Field

[0001] The utility model relates to the technical field of gate machines, in particular to a gate machine based on the combination of a camera and a speed sensor. Background Art

[0002] With the rapid development of artificial intelligence and intelligent manufacturing technologies, smart access gates are widely used in many public places, such as airports, high-speed rail stations, parks, etc., bringing great convenience to personnel control and passage in public areas.

[0003] One of the most important functions of the intelligent access gate is to control the number of people passing through while ensuring safety, accurately count the number of people passing through, and ensure that no pedestrians are caught, and to prevent tailgating and other situations. In order to achieve this goal, the common methods currently used are to solve the problem through the infrared grating in the gate channel and the visual (camera) solution. Among them, the infrared grating solution is to determine whether there is an obstruction in the channel to determine whether there is a person passing through. This solution cannot accurately distinguish whether it is a person or other items such as luggage, especially for children and long luggage. The detection effect is very poor, and it is easy to catch people after a wrong judgment; while the visual solution generally uses the artificial intelligence detection method of the human body / human head to determine whether there is a person passing through the channel. This solution is easy to mistake hats, large backpacks and other items for human bodies, causing false alarms of tailgating and inaccurate counting. In addition, there are also gate products that combine infrared gratings and visual solutions, but they are also unable to accurately distinguish between people and objects. Hats, human sculptures or hair-like items placed on suitcases are easily misjudged as pedestrians, causing false alarms of tailgating and inaccurate counting of passers-by.

[0004] In view of this, the designer of the present invention has made in-depth considerations on the many deficiencies and inconveniences caused by the inability of existing smart gates to accurately distinguish between people and objects, and has actively researched, improved and trial-produced the present utility model. Utility Model Content

[0005] The purpose of the utility model is to overcome the deficiencies of the prior art and provide a gate machine based on the combination of a camera and a speed sensor which can improve the passage efficiency and safety of the gate machine.

[0006] In order to achieve the above purpose, the solution of the utility model is:

[0007] A gate machine based on the combination of a camera and a speed sensor, which includes two columns arranged at intervals, a gate passage is formed between the two columns, an openable gate is installed in the gate passage, the gate is connected to a control system, a camera and a speed sensor are installed above the gate passage to form a three-dimensional image of image + point cloud, the control system includes a calibration module, a detection and tracking module, a judgment module and a gate control module, the calibration module is used to align and calibrate the camera and the speed sensor, the detection and tracking module is connected to the camera, and is used to detect and track a head or a human body in a video or image obtained by the camera; the judgment module is connected to the detection and tracking module and the speed sensor, and is used to determine whether the detection area is a human body or other objects; the gate control module is connected to the gate, and is used to control the gate switch.

[0008] Furthermore, the speed sensor is any one of millimeter wave radar, laser radar or infrared.

[0009] Furthermore, the control system also includes an alarm module, and the alarm module is connected to the judgment module.

[0010] Furthermore, the control system also includes a counting module, which is connected to the judgment module to count the number of passers-by.

[0011] Furthermore, a support is installed on the top surface of the column, and the camera and the speed sensor are both arranged on the top of the support and face the gate passage.

[0012] After adopting the above scheme, the utility model combines the image video taken by the camera with the point cloud map formed by the speed sensor through the gate machine based on the combination of camera and speed sensor. On the basis of visual detection of the human body, combined with the point cloud image of the speed sensor, the dynamic Doppler effect is used to accurately distinguish whether the visually detected part is a real human body or other objects, accurately calculate the number of people passing through, reduce or even eliminate false alarms of tailgating, and improve the passage efficiency and safety of the gate machine.

[0013] Compared with the prior art, the gate machine based on the combination of camera and speed sensor in the utility model has the following advantages:

[0014] 1. The utility model combines vision and millimeter-wave radar sensors, retains visual information, inherits the ability of the visual solution to distinguish between people and objects to a certain extent, and is more accurate than the traditional infrared grating solution.

[0015] 2. The utility model adopts a method combining vision and millimeter-wave radar sensors. Through the millimeter-wave radar point cloud information, it makes up for the shortcomings of the original vision scheme or vision + infrared grating scheme that objects similar to human heads such as hats and large backpacks are easily mistaken for human bodies, reducing false alarms of tailgating and lowering the error rate of pass counting.

[0016] 3. Based on vision, the utility model adds millimeter-wave radar point cloud, and through the dynamic Doppler effect, accurately determines whether the human body detected by vision is a real human body or other objects. Compared with other solutions and technologies, it can accurately distinguish between people and objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural schematic diagram of the utility model.

[0018] Figure 2 It is a schematic diagram of the control system of the utility model.

[0019] Figure 3 The utility model is a flow chart of the gate passage. DETAILED DESCRIPTION

[0020] In order to further explain the technical solution of the present utility model, the present utility model is described in detail below through specific embodiments.

[0021] like Figure 1 and Figure 2 As shown, the utility model discloses a gate machine based on the combination of a camera and a speed sensor, which includes two columns 1 and a control system 3 arranged at intervals, a gate passage 11 is formed between the two columns 1, an openable gate 2 is installed in the gate passage, the gate 2 is connected to the control system 3, a camera 4 and a speed sensor 5 are installed above the gate passage 11, the speed sensor in the utility model is a millimeter wave radar, which forms a three-dimensional image of image + point cloud, in this embodiment, a pillar 12 is installed on the top surface of the column 1, the camera 4 and the speed sensor 5 are both arranged at the top of the pillar 12 and facing the gate passage 11. The control system 3 includes a calibration module 31, a detection and tracking module 32, a judgment module 33, a gate control module 34, a counting module 35 and an alarm module 36. The calibration module 31 is used to align and calibrate the camera 4 and the speed sensor 5. The detection and tracking module 32 is connected to the camera 4 to detect and track the head or human body in the video or image obtained by the camera 4; the judgment module 33 is connected to the detection and tracking module 32 and the speed sensor 5 to determine whether the detection area is a human body or other objects; the gate control module 34 is connected to the gate 2 to control the opening and closing of the gate 2.

[0022] like Figure 3 As shown, the passage counting method of the gate machine based on the combination of a camera and a speed sensor of the utility model is:

[0023] S1. Align and calibrate the camera 4 and the millimeter wave radar 5 to form a three-dimensional image of image + point cloud. The calibration method may be an existing common calibration method;

[0024] S2. During each opening and closing process of the gate 2, the head and body are detected and tracked in the video or image obtained by the camera 5. The detection and tracking of the head and body can be based on the existing common deep learning detection and tracking algorithms such as YOLO, which is the existing technology and is not limited by the present utility model;

[0025] S3, according to the human head and human body position frame in the image, correspond to the millimeter wave point cloud map, and combine the point cloud maps of a period of time before and after, use the dynamic Doppler effect to determine whether this part of the area is a human body or other objects;

[0026] S4, the application layer counts the number of people passing through based on the judgment result of S3 and notifies the gate control module to perform corresponding operations or alarms.

[0027] It has been observed that the human body is a non-rigid body. When moving, the movement directions and speeds of different parts of the body are not consistent due to common movements such as leg and arm swinging, head shaking, and changes in clothing folds caused by movement. However, the movement directions and speeds of various parts of these luggage or other items that are difficult to distinguish visually remain consistent with those of the whole body when the person is moving.

[0028] Millimeter wave radar is a common speed measurement solution and is widely used. It generally uses the Doppler effect to measure speed. Therefore, the main idea of ​​the utility model is to use the dynamic Doppler effect on the basis of the visual solution to judge whether the movement direction and speed of each part of the human body detected by vision are consistent with the whole through its corresponding millimeter wave radar point cloud. By adding a speed judgment, it can more accurately distinguish whether it is a person or an object, reducing the false alarm of the visual solution.

[0029] The utility model combines a camera with a millimeter-wave radar for gate passage counting, utilizing the phenomenon that a person is a flexible object, and when walking, the movement speeds and directions of various positions of the body are inconsistent, while each component or position of other objects such as luggage is consistent with the overall movement speed and direction. The dynamic Doppler effect of millimeter-wave imaging is used to calculate the movement speed and direction of each position, and distinguish between people and other objects.

[0030] A simple visual solution can only distinguish some people and objects. Luggage items that look like human heads or / and human bodies are easily mistaken for people, resulting in false detections and false alarms. The present invention adds a millimeter-wave radar module, which can accurately distinguish people and objects based on the visual solution. A simple millimeter-wave radar can only measure speed and cannot determine whether each part belongs to the same person or object. This part can be completed by the visual solution, so the combination of vision + millimeter-wave radar proposed in the present invention can achieve perfect results.

[0031] The millimeter wave radar in the present invention may also be replaced by other 3D measurement sensors, such as laser radar, infrared, etc. The above embodiments and drawings do not limit the product form and style of the present invention, and any appropriate changes or modifications made by ordinary technicians in the technical field shall be deemed to be within the patent scope of the present invention.

Claims

1. A gate based on a combination of a camera and a speed sensor, characterized in that: include: Two columns are arranged at intervals, and a gate passage is formed between the two columns. A gate that can be opened and closed is installed in the gate passage, and the gate is connected to the control system. A camera and a speed sensor are installed above the gate passage to form a three-dimensional image of image + point cloud. The control system includes a calibration module, a detection and tracking module, a judgment module and a gate control module. The calibration module is used to align and calibrate the camera and the speed sensor. The detection and tracking module is connected to the camera to detect and track the head or body in the video or image obtained by the camera; The judgment module is connected to the detection and tracking module and the speed sensor to determine whether the detection area is a human body or other objects; the gate control module is connected to the gate to control the gate switch.

2. The gate based on the combination of camera and speed sensor according to claim 1 is characterized in that: The speed sensor is any one of millimeter wave radar, laser radar or infrared.

3. The gate based on the combination of camera and speed sensor as claimed in claim 1, characterized in that: The control system further comprises an alarm module, and the alarm module is connected to the judgment module.

4. The gate based on the combination of camera and speed sensor as claimed in claim 1, characterized in that: The control system also includes a counting module, which is connected to the judging module to count the number of passers-by.

5. The gate based on the combination of camera and speed sensor as claimed in claim 1, characterized in that: A support is installed on the top surface of the column, and the camera and the speed sensor are both arranged on the top of the support and face the gate passage.

Images